Search

The existence of information and communication technology in this digital era has changed the way people interact with money and finance, especially with the emergence of internet developments such as mobile banking, e-commerce, and fintech which have brought great progress in the way people transact and manage finances. Financial transactions that are increasingly related to digital technology bring new benefits and challenges for UMKM business actors, the introduction of digital financial literacy in the context of UMKM is an important step in facing technology-driven business transformation. However, until now there are still many UMKM who still lack understanding of digital financial literacy. Lack of understanding about online transaction security, selection of appropriate services, and digital data management can hinder growth and potentially increase risks for UMKM later. This makes the need for digital financial literacy for UMKM even more real, this research will focus on helping to provide understanding to UMKM actors, especially Medan City about digital financial literacy through interactive educational applications by applying the User Centered Design Method for the development of their user interface, it is hoped that through this application UMKM actors in the city of Medan can optimize their understanding of digital financial literacy so that they can Take full advantage of financial technology to improve operational efficiency, grow businesses, and participate in the digital economy more effectively.

Flooding is one of the most common natural disasters in Indonesia, including in Merbau Sub-district, North Labuhan Batu Regency, North Sumatra. This disaster can cause huge losses, both material and non-material. One of the efforts to reduce the impact of flooding is to immediately distribute aid to affected communities. The distribution of post-flood aid requires careful planning so that it can run quickly and on target. In this research, we propose the use of greedy algorithm to support the decision of choosing post-flood aid distribution route in Merbau District. The greedy algorithm is an algorithm that chooses the best solution at each step, hoping to find the optimal solution as a whole. This research was conducted using data on the location of disaster points, the location of aid distribution points, and the distance between points. The results showed that the greedy algorithm can produce a faster aid distribution route compared to the conventional route.

Seventy-two percent of the Muslim population in Indonesia is illiterate in Arabic script, thus requiring early childhood education. Young children often struggle to directly learn Arabic script, necessitating the role of instructors. However, most instructors in madrasahs do not teach Arabic script intensively, prompting the need for interactive and engaging learning media for children. Therefore, the development of an Android application for teaching Arabic script to young children is crucial. By employing the Rapid Application Development (RAD) method, an Android application can be developed, incorporating gamification for interactive learning and the AI Life Cycle for developing a 3D Arabic script object detection model useful for automatic correction systems. The AI model is developed using SSD MobileNet v2 FPN Lite 640x640. Sixteen trials are conducted for separate letter modeling and two for connected letters. The separate letter modeling yields the best result in the 8th trial with a total loss of 0.182882 out of 1, using parameters of 16 batch size, 30,000 num steps, and checkpoint from the 7th trial. Connected letter modeling yields the best result in the 2nd trial with parameters of 8 batch size, 14,000 num steps, and checkpoint from the 1st trial. The output is an offline Android application integrating gamification and AI technology, featuring two categories of questions, each with 20 items.

PT. Neora Infrastructure Indonesia is an industrial company engaged in telecommunications and/or telephone transmission network construction services, collaborating with Fiberstar. This necessitates the company to constantly require competent human resources. However, the company must compete with others to acquire competent personnel. Nevertheless, the employee selection process at PT. Neora Infrastructure Indonesia is still manual and takes 2 - 4 weeks, making it difficult for the company to compete in acquiring competent personnel. Therefore, it was decided to develop a decision support system to assist the company in efficiently and effectively selecting employees. The decision support system for employee selection uses the ROC and CPI methods. The ROC method is used to determine the criteria weight values, while the CPI method is used for ranking all alternatives. Quantitative method is the research method used, and several types of UML diagrams including Activity diagram, Class diagram, and Use case diagram are utilized for system design. The RAD method is employed for system development. With the establishment of the decision support system for employee selection using the ROC and CPI methods, it is expected that the company can compete with others in obtaining competent human resources and that the employee selection process becomes efficient and effective

The retail industry has undergone a significant transformation. Jumbo Mart is one of the retail businesses that consistently meet customer needs in Pekanbaru city. The number of similar businesses makes Jumbo Mart need to have a competitive advantage to compete. One of them is by improving the shopping experience, especially in the aspect of goods layout. However, based on observations and interviews, the determination of the layout of goods at Jumbo Mart is still not optimal. The number of products and variety of items sold is one of the reasons. In addition, customer transaction history has not been used further and is only stored as an archive of monthly reports. Whereas analyzing the data can be an effective solution in arranging the layout of goods based on customer behavior. Therefore, this research proposes a data mining technique with the FP-Growth algorithm to find association rules between categories of goods. The implementation results with a minimum support value of 15% and confidence of 50% get 7 association rules, but only 6 are proven valid. The valid association rules are then proposed to be placed close together so that it can make it easier for buyers to find the desired items. In addition, items with the consumption category are dominant as consequents, which can be found in the four association rules. This indicates that items with consumption categories have great appeal and are an important part of customers' shopping behavior and habits.

Most people, especially women, have a great desire to have white, healthy, clean and well-maintained facial skin. However, their knowledge about facial skin types is still limited, even though consulting with an expert requires a lot of time and money which results in someone not paying attention to facial skin type when carrying out treatment. Therefore, an expert system is needed that can help identify facial skin types. A rule base is a rule created based on expert knowledge needed to create an expert system. The forward chaining method is a search method or forward tracing technique that starts from existing information and combines rules to produce a conclusion or goal. The research results show that this application can run well and is suitable for use. Based on the results of system testing from an expert, it was concluded that identifying facial skin types based on facial skin criteria using the forward chaining method had an accuracy rate of 84% where the results of system testing produced several conclusions about the appropriate type of facial skin with the selected criteria data.

The leading cause of human death is heart disease (HD) worldwide which generally occurs when the heart is unable to push enough fresh, oxidized blood to the rest of the body. This disease makes it more difficult for the heart muscle to pump blood efficiently and causes chest pain, chest pressure, shortness of breath, pain in the neck and jaw. The aim of this research is to compare and obtain the best accuracy results from the three methods used, namely Random Forest, Extra Trees Classifier and Naïve Bayes. The results of this research prove that the Extra Trees Classifier method with an accuracy of 86.93% has higher results compared to the Naïve Bayes method with an accuracy of 84.21%, and the Random Forest Classifier method with an accuracy of 84.21%. Meanwhile, the AUC results obtained by the Extra Trees Classifier method are higher than other methods with an AUC of 93.81%.

Technological advances are developing just like scientific advances. One example of the latest technological advances is the online attendance system. At MTsN Binjai, attendance currently uses electronic-based technology, namely fingerprints, but still has several weaknesses. For example, the device cannot be used when the electricity fails, and the attendance process must be carried out alternately due to dependence on one device. In addition, fingerprints only record attendance, do not facilitate the submission of permits or illnesses, so the admin must do double work. The thing that was researched was the use of descriptive methods and the development of a system with the prototype method. The purpose of this study is to generate an Android-based absence application that can be accessed through the personal cellphones of teachers and staff who can facilitate attendance and submission of sick permits. The results of the study are expected to be a reference for schools to improve discipline for attendance.

This paper reports on a scientometric analysis bolstered by human-in-the-loop, domain experts, to examine the field of metal-organic frameworks (MOFs) research. Scientometric analyses reveal the intellectual landscape of a field. The study engaged MOF scientists in the design and review of our research workflow. MOF materials are an essential component in next-generation renewable energy storage and biomedical technologies. The research approach demonstrates how engaging experts, via human-in-the-loop processes, can help develop a comprehensive view of a field’s research trends, influential works, and specialized topics.

This study investigated the publication behaviour of 573 chief editors managing 432 Social Sciences journals in Turkey. Direct inquiries into editorial qualifications are rare, and this research aims to shed light on editors’ scientific leadership capabilities.

The quantitative rankings of over 55,000 institutions and their institutional programs are based on the individual rankings of approximately 30 million scholars determined by their productivity, impact, and quality.

Large Language Models (LLMs), exemplified by ChatGPT, have significantly reshaped text generation, particularly in the realm of writing assistance. While ethical considerations underscore the importance of transparently acknowledging LLM use, especially in scientific communication, genuine acknowledgment remains infrequent. A potential avenue to encourage accurate acknowledging of LLM-assisted writing involves employing automated detectors. Our evaluation of four cutting-edge LLM-generated text detectors reveals their suboptimal performance compared to a simple ad-hoc detector designed to identify abrupt writing style changes around the time of LLM proliferation. We contend that the development of specialized detectors exclusively dedicated to LLM-assisted writing detection is necessary. Such detectors could play a crucial role in fostering more authentic recognition of LLM involvement in scientific communication, addressing the current challenges in acknowledgment practices.

A text generation based multidisciplinary problem identification method is proposed, which does not rely on a large amount of data annotation.

The disseminating of academic knowledge to nonacademic audiences partly relies on the transition of subsequent citing papers. This study aims to investigate direct and indirect impact on technology and policy originating from transformative research based on ego citation network.

Abstract As technology continues to advance, the integration of generative artificial intelligence tools in various sectors, including education, has gained momentum. ChatGPT, an extensively recognized language model created by OpenAI, has gained significant importance, particularly in education. This study investigates the awareness, acceptance, and adoption of ChatGPT, a state-of-the-art language model developed by OpenAI, in higher education institutions across China. This study applies the partial least squares structural equation modeling (PLS-SEM) method for examining data collected from 320 Chinese university students. The study’s conceptual framework integrates key determinants from the Technology Acceptance Model (TAM) and extends it by incorporating perceived intelligence as a critical factor in the adoption process. The study findings reveal that ChatGPT awareness significantly influences the intention to adopt ChatGPT. Perceived ease of use, usefulness, and intelligence significantly mediate the association between ChatGPT awareness and adoption intention of ChatGPT. Additionally, perceived trust significantly moderates the relationship between ChatGPT awareness and perceived ease of use, usefulness, and intelligence. Moving forward, in order to maintain students’ critical thinking skills and inventiveness in their assessment writing, assessments must promote the safe use of ChatGPT. Therefore, educators will be crucial in ensuring that artificial intelligence tools are used in assessments ethically and suitably by providing clear guidelines and instructions.

Abstract Machine learning is becoming a preferred method for the virtual screening of organic materials due to its cost-effectiveness over traditional computationally demanding techniques. However, the scarcity of labeled data for organic materials poses a significant challenge for training advanced machine learning models. This study showcases the potential of utilizing databases of drug-like small molecules and chemical reactions to pretrain the BERT model, enhancing its performance in the virtual screening of organic materials. By fine-tuning the BERT models with data from five virtual screening tasks, the version pretrained with the USPTO–SMILES dataset achieved R2 scores exceeding 0.94 for three tasks and over 0.81 for two others. This performance surpasses that of models pretrained on the small molecule or organic materials databases and outperforms three traditional machine learning models trained directly on virtual screening data. The success of the USPTO–SMILES pretrained BERT model can be attributed to the diverse array of organic building blocks in the USPTO database, offering a broader exploration of the chemical space. The study further suggests that accessing a reaction database with a wider range of reactions than the USPTO could further enhance model performance. Overall, this research validates the feasibility of applying transfer learning across different chemical domains for the efficient virtual screening of organic materials. Scientific contribution This study verifies the feasibility of applying transfer learning to large language models in different chemical fields to help organic materials perform virtual screening. Through the comparison of transfer learning from different chemical fields to a variety of organic material molecules, the high precision virtual screening of organic materials is realized.

Abstract Every year, more than 19 million cancer cases are diagnosed, and this number continues to increase annually. Since standard treatment options have varying success rates for different types of cancer, understanding the biology of an individual's tumour becomes crucial, especially for cases that are difficult to treat. Personalised high-throughput profiling, using next-generation sequencing, allows for a comprehensive examination of biopsy specimens. Furthermore, the widespread use of this technology has generated a wealth of information on cancer-specific gene alterations. However, there exists a significant gap between identified alterations and their proven impact on protein function. Here, we present a bioinformatics pipeline that enables fast analysis of a missense mutation’s effect on stability and function in known oncogenic proteins. This pipeline is coupled with a predictor that summarises the outputs of different tools used throughout the pipeline, providing a single probability score, achieving a balanced accuracy above 86%. The pipeline incorporates a virtual screening method to suggest potential FDA/EMA-approved drugs to be considered for treatment. We showcase three case studies to demonstrate the timely utility of this pipeline. To facilitate access and analysis of cancer-related mutations, we have packaged the pipeline as a web server, which is freely available at https://loschmidt.chemi.muni.cz/predictonco/ . Scientific contribution This work presents a novel bioinformatics pipeline that integrates multiple computational tools to predict the effects of missense mutations on proteins of oncological interest. The pipeline uniquely combines fast protein modelling, stability prediction, and evolutionary analysis with virtual drug screening, while offering actionable insights for precision oncology. This comprehensive approach surpasses existing tools by automating the interpretation of mutations and suggesting potential treatments, thereby striving to bridge the gap between sequencing data and clinical application.

Abstract Mass spectral libraries have proven to be essential for mass spectrum annotation, both for library matching and training new machine learning algorithms. A key step in training machine learning models is the availability of high-quality training data. Public libraries of mass spectrometry data that are open to user submission often suffer from limited metadata curation and harmonization. The resulting variability in data quality makes training of machine learning models challenging. Here we present a library cleaning pipeline designed for cleaning tandem mass spectrometry library data. The pipeline is designed with ease of use, flexibility, and reproducibility as leading principles. Scientific contribution This pipeline will result in cleaner public mass spectral libraries that will improve library searching and the quality of machine-learning training datasets in mass spectrometry. This pipeline builds on previous work by adding new functionality for curating and correcting annotated libraries, by validating structure annotations. Due to the high quality of our software, the reproducibility, and improved logging, we think our new pipeline has the potential to become the standard in the field for cleaning tandem mass spectrometry libraries. Graphical Abstract

Abstract Motivation Chemical space embedding methods are widely utilized in various research settings for dimensional reduction, clustering and effective visualization. The maps generated by the embedding process can provide valuable insight to medicinal chemists in terms of the relationships between structural, physicochemical and biological properties of compounds. However, these maps are known to be difficult to interpret, and the ‘‘landscape’’ on the map is prone to ‘‘rearrangement’’ when embedding different sets of compounds. Results In this study we present the Hilbert-Curve Assisted Space Embedding (HCASE) method which was designed to create maps by organizing structures according to a logic familiar to medicinal chemists. First, a chemical space is created with the help of a set of ‘‘reference scaffolds’’. These scaffolds are sorted according to the medicinal chemistry inspired Scaffold-Key algorithm found in prior art. Next, the ordered scaffolds are mapped to a line which is folded into a higher dimensional (here: 2D) space. The intricately folded line is referred to as a pseudo-Hilbert-Curve. The embedding of a compound happens by locating its most similar reference scaffold in the pseudo-Hilbert-Curve and assuming the respective position. Through a series of experiments, we demonstrate the properties of the maps generated by the HCASE method. Subjects of embeddings were compounds of the DrugBank and CANVASS libraries, and the chemical spaces were defined by scaffolds extracted from the ChEMBL database. Scientific contribution The novelty of HCASE method lies in generating robust and intuitive chemical space embeddings that are reflective of a medicinal chemist’s reasoning, and the precedential use of space filling (Hilbert) curve in the process. Availability https://github.com/ncats/hcase Graphical Abstract

In the face of rapid technological advancements and evolving consumer demands, businesses increasingly turn to digitalization to remain competitive and redefine industry standards. This article delves into the crucial role of enterprise integration within the context of digital transformation, with a specific focus on manufacturing Small and Medium Enterprises (SMEs). Based on fieldwork within the Innovation Factory North project in Denmark, the study presents a three-dimensional framework – architecture, capabilities, and governance – that is instrumental in guiding manufacturing SMEs through the complexities of digitalization. The architecture dimension explores the dichotomy between monolithic and fog (edge) architectures, highlighting their respective merits and challenges. The capabilities dimension introduces the concept of operational capability stacking, demonstrating how specific information systems, when integrated, yield new capabilities. A list of common capabilities in manufacturing enterprises is provided, offering inspiration for manufacturing SMEs to enhance their operational efficiency. The governance dimension injects a strategic dimension into the digitalization process, emphasizing the need for a balanced approach between technology-driven and process-driven optimization methodologies. The research underscores the strategic nature of digitalization in manufacturing SMEs, recognizing the interconnectedness of architecture, capabilities, and governance in achieving sustained success in the digital landscape. The proposed three-dimensional framework is substantiated through real-world applications within Denmark’s Innovation Factory North project, providing tangible insights and scenarios to navigate the intricacies of the digital landscape effectively.

Case Management Modeling and Notation (CMMN) and iStar are two distinct, multi-purposed modeling techniques that may be used to represent organizational challenges at separate levels of abstraction. CMMN, a flexible process-oriented technique, aims to extract knowledge that enhances the representational capacity of activity flows for a specific case. Conversely, the iStar framework adopts a goal-oriented modeling approach, effectively capturing the interplay of social actors and their influence on the attainment of organizational objectives. While prior studies have explored methods for integrating these techniques to attain a more comprehensive understanding of the organizational landscape, these remain primarily associated with a distinct level of the organizational matrix (i.e., CMMN for operations and tactics, and iStar for more strategic aspects). As such, any effort to evaluate their semantic proximity appears fragmented, as it deals only with the partial association of specific notations and elements. This article describes the conduct of a dual-purposed literature review to identify specific criteria that might accommodate a more holistic assessment of the two modeling techniques; these criteria are then employed as a guiding framework to construct a set of propositions that articulate the areas of complementarity and/or divergence between these two techniques, as identified in previous research. These propositions are subsequently subjected to validation by domain experts, leveraging a real-world case study in the educational domain. The results show that there can be areas of semantic convergence between the two techniques, suggesting their parallel use to effectively model complex domain problems. Overall, the present study aims to crystallize an approach for conducting complex modeling comparisons that transcends technical considerations.

With the latest advances in natural language processing technologies, multi-participant conversation summarization features are now embedded in the most widely used collaboration platforms offered by industry leaders such as Microsoft, Google, and Zoom. This allows employees to streamline their work and increase efficiency by summarizing long chat threads. In this study, an attempt has been made to perceive summarized chat conversations as a tool for knowledge discovery and reusable information extraction within an organization in general or during projects. To this end, recent scientific articles have been reviewed to identify the most effective techniques and approaches for summarizing chat threads and conversations alongside the challenges and peculiarities of collaborative text-based communication. In addition, significant attention has been paid to the further utilization of the extracted information to represent the knowledge for further reuse.

Managing and dealing with variability in business processes and the IT landscape is a common challenge in the everyday practice of most enterprises and organizations. Recent studies have observed that digital transformation, Internet-of-Things solutions and the introduction of artificial intelligence cause changes and challenges in enterprises that simultaneously require variability on several levels, for instance, business processes, data architecture, and services. Enterprise architecture models are considered a suitable way to visualize and manage dependencies between different levels of an enterprise. However, the management of variability in enterprise architectures has not received much attention in scientific research. This article aims to contribute to a better understanding of future investigation needs. Using a systematic literature analysis, the article structures the existing research work in the field and examines real-world challenges of variability based on a case study. We argue that there is a need for new constructs in enterprise architecture models that allow for expressing dependencies between variations on different enterprise architecture layers.

Data analytics projects have become a common accomplishment in many enterprises. However, establishing a data analytics project requires consideration of many factors that are not always recognized at the very beginning of the project. This study seeks to identify what generic requirements must be defined for data analytics projects and what analytics project attributes need to be addressed by these requirements. It proposes a requirements template for the generic requirements of the analytics projects. The template is intended to be used to reduce the complexity of starting the analytics projects by providing a checklist of requirements to be considered at the beginning of the project. The template is derived from analyzing 16 data analytics project reports for descriptive, diagnostic, predictive, and prescriptive analytics tasks. The template is then validated by analyzing its compliance with 20 analytics projects within the real estate domain using the corresponding research articles.

The intricacies of relay-aided device-to-device (D2D) communication within the novel fluctuating two-ray (FTR) model is inspected in this study. Two cases with respect to co-channel interference are considered here, in case A, co-channel interferences due to unmanned aerial vehicles affect the relay while at the D2D receiver, the co-channel interferences follow the FTR model. In case B, interferences due to unmanned aerial vehicles affect the D2D receiver and the relay now experiences interferences over the FTR channels. The channel of a co-channel interference due to unmanned aerial vehicle is assumed to be a recently proposed single-shadowing based composite fading model. Further enhancing operational efficiency, the D2D receiver incorporates a power splitting mechanism for power harvesting. In this analysis a characteristic function-based approach is applied, which serves as the foundational cornerstone for deriving expressions that govern the outage and success probabilities. These intricate expressions are connected with critical factors, such as the dynamics of path-loss phenomena, the spatial separation between the interferers and the relay, the varying distances between the interferers and the D2D receiver, and the physical gap bridging the relay and the D2D receiver. Building upon the derived expressions, an extensive numerical analysis unfurls them for both interference cases.

This paper describes a 2-stage tactic for achieving an odd-order variable-bandwidth (OO-VBW) filter with absolutely guaranteed stability. The design methodology aims to minimize the p-norm amplitude-response error while maintaining the OO-VBW-filter's stability. In order to tune OO-VBW filter's amplitude response, we utilize a kind of functions of the bandwidth (BW)-tuning parameter to express the filter's coefficients. Because those functions have changeable function values, the designed OO-VBW-filter's amplitude response possesses variability. Another important concern in this paper is the stability issue. When the filter coefficient values expressed as function values are changed, the OO-VBW filter that has feedback structures may become unstable. This necessitates that the OO-VBW-filter's stability must always be preserved during real-time tuning. In order to stabilize such an OO-VBW filter that has feedback structures, a coefficient-conversion (CC) algorithm is adopted, and it is incorporated into the 2-stage methodology for obtaining the OO-VBW filter. To illustrate both the stability and high approximation accuracy, the design of an OO-VBW bandpass filter is simulated. The simulation details verify that both stability and high accuracy can be successfully achieved.

The aim of the article is to present a model of decision-making processes in the design of flow-based Internet of Things systems. This is an attempt to fill the research gap in the construction of decision-making systems for designing both infrastructure and the application development environment. Therefore, the aim of the article is to show a decision-making model for selecting infrastructure and design environment for the development of IoT systems. In order for these processes to be implemented, it is necessary to use containerization processes and Flow Based Programming systems, both at the stage of system prototyping and its subsequent implementation.The model verification was carried out on the example of design in accordance with the assumptions of the parallel design process model proposed in the article, using the infrastructure as code approach in infrastructure design, the Docker process for creating containers, publisher and subscriber design in Node Red, and Postgress database design.

The performance of the cognitive non-orthogonal multiple access (NOMA) systems with random locations is examined in the present paper. More precisely, the positions of the secondary receivers (SRs) are randomly distributed inside a disk with the centre at the secondary transmitter (ST). The primary networks (PNs), on the other hand, are not necessarily inside the transmission range of the ST. Two user selection schemes namely, path-loss-based and channel-gain-based approaches, are adopted to serve two SRs according to the power domain NOMA technique. Additionally, the impact of the imperfect hardware impairments (HIs) on both transmitters and receivers is also taken into consideration. Furthermore, a recently proposed transmit power allocation at the ST is employed to concurrently maximize the system performance of the secondary networks (SNs) and to strictly protect the quality-of-service (QoS) of the PNs. The correctness of the developed mathematical frameworks is corroborated by simulation results based on the Monte Carlo method. Finally, we unveil that there is no scheme that always outperforms another. Thus, an adaptive scheme can be proposed to optimize the networks' performance. Our findings also reveal that the outage probability (OP) under the harmful effect of HI can be effectively mitigated by increasing the transmit power of the ST. Additionally, increasing the transmit power of the PT is also beneficial for the SNs with the adopted transmit power allocation at the ST.

This paper presents a Machine Learning and IoT-based intelligent medical system for the detection and monitoring of patient stress. This system is made up of a medical kit measuring the oxygen saturation, the heart rate and the galvanic skin response thanks to sensors attached at the top of the patient’s hand which send the measured physiological values to the Firebase server. A voting classifier, combining five Machine Learning algorithms (Logistic Regression, K-Nearest Neighbors, Support Vector Machine, Decision Tree and Random Forest) using holdout and K-fold cross-validation, was implemented on a Raspberry board installed in the doctor’s office. The proposed system can make predictions with the Soft Voting classifier with an accuracy that reaches 78%.

Due to their high price, RFID tags are not yet widely used in applications for monitoring, tracing, and information inscribing, such as manufacturing lines, supply chains, and inventory management. The currently utilized barcode-like tracing techniques lack security, experience data limitation, are semi-automated, and are prone to external damage. Chipless RFID technology is a potential replacement, able to overcome shortcomings. However, due to the uncertainty created by the absence of a communication antenna and microchip, the detection of chipless RFID tags has a 2–5% detection error rate. The recent advancement in spatial chipless RFID technology has opened up a plethora of opportunities to use deep learning, which is becoming increasingly popular, to detect chipless RFID tags from a 2D virtual image created from the backscattered signal. In this study, we presented a comprehensive and exhaustive investigation of several deep learning techniques to improve the detection capability of deep learning-based chipless RFID tag detection. Our proposed fine-tuning methods yielded better mathematical metrics on the state-of-the-art on its original setup. A confusion matrix used to measure error rates on unseen data shows that the improved deep learning architecture we introduced works at roughly 99.99% accuracy.

Data Envelopment Analysis is applied to a sample of IT projects. The application is based on a discussion concerning the choice of inputs and outputs in the case when the units being evaluated are projects. This discussion takes into account various understandings of project success and thus project desired outputs, as well as the ambiguity and a strong situation dependence concerning the choice of project inputs, understood as variables that should be minimized. The sample consists of over 80 IT projects implemented in Poland. The applied DEA model has permitted to identify the projects that clearly stand out from the rest of the sample as less efficient. Their analysis made it possible to draw conclusions on the choice of inputs and outputs. A set of indications on the context-specific choice of inputs and outputs were formulated.

Abstract Compact and pole-controllable resonators and bandpass filter (BPF) using a composite right/left-handed transmission line (CRLH-TL) are designed in this study. The distributed constant line in the CRLH-TL has applied a tap-coupling technique, and one tap-coupled stub is loaded with the left-handed (LH) circuit. Attenuation poles are generated when the input susceptance of the stub diverges. In the tap-coupled CRLH-TL resonator, the attenuation pole is controlled at either the desired lower or higher region frequency than a resonant frequency by adjusting the circuit parameters. Also, the BPF constructed by the CRLH-TL resonators is designed based on a filter design theory, where the attenuation poles are located at lower and higher region frequencies than a negative-first-order frequency. The BPF with microstrip structure is fabricated using MEGTRON6 R-5775 ( $$\varepsilon _\text {r}$$ ε r : 3.7, h: 0.63 mm, t: 18 μm), chip capacitors, and wire inductors. The simulated results show that the desired characteristics are approximately satisfied, i.e., we can design the CRLH-TL BPF which is controllable for the attenuation poles at both lower and higher region frequencies than a resonant frequency. The measured results are good agreement with the simulation. The negative-first-order frequency is generated at 2.00 GHz with 133 MHz bandwidth, i.e., the fractional bandwidth is 6.65 %. The resonator lengths in the BPF are shortened by 81.5 % and 75.0 % in comparison with a conventional half-wavelength ( $$\lambda /2$$ λ / 2 ) openstub, and the size of the fabricated BPF is 0.18 $$\lambda _\text {g} \times$$ λ g × 0.17 $$\lambda _\text {g}$$ λ g . Therefore, a compact BPF with two controllable attenuation poles is realized by the tap-coupled CRLH-TL.

Abstract Objective This study was designed to develop and validate a robust predictive model for one-year mortality in elderly coronary heart disease (CHD) patients with anemia using machine learning methods. Methods Demographics, tests, comorbidities, and drugs were collected for a cohort of 974 elderly patients with CHD. A prospective analysis was performed to evaluate predictive performances of the developed models. External validation of models was performed in a series of 112 elderly CHD patients with anemia. Results The overall one-year mortality was 43.6%. Risk factors included heart rate, chronic heart failure, tachycardia and β receptor blockers. Protective factors included hemoglobin, albumin, high density lipoprotein cholesterol, estimated glomerular filtration rate (eGFR), left ventricular ejection fraction (LVEF), aspirin, clopidogrel, calcium channel blockers, angiotensin converting enzyme inhibitors (ACEIs)/angiotensin receptor blockers (ARBs), and statins. Compared with other algorithms, an ensemble machine learning model performed the best with area under the curve (95% confidence interval) being 0.828 (0.805–0.870) and Brier score being 0.170. Calibration and density curves further confirmed favorable predicted probability and discriminative ability of an ensemble machine learning model. External validation of Ensemble Model also exhibited good performance with area under the curve (95% confidence interval) being 0.825 (0.734–0.916) and Brier score being 0.185. Patients in the high-risk group had more than six-fold probability of one-year mortality compared with those in the low-risk group (P

Abstract The detection of hate speech (HS) in online platforms has become extremely important for maintaining a safe and inclusive environment. While significant progress has been made in English-language HS detection, methods for detecting HS in other languages, such as Bengali, have not been explored much like English. In this survey, we outlined the key challenges specific to HS detection in Bengali, including the scarcity of labeled datasets, linguistic nuances, and contextual variations. We also examined different approaches and methodologies employed by researchers to address these challenges, including classical machine learning techniques, ensemble approaches, and more recent deep learning advancements. Furthermore, we explored the performance metrics used for evaluation, including the accuracy, precision, recall, receiver operating characteristic (ROC) curve, area under the ROC curve (AUC), sensitivity, specificity, and F1 score, providing insights into the effectiveness of the proposed models. Additionally, we identified the limitations and future directions of research in Bengali HS detection, highlighting the need for larger annotated datasets, cross-lingual transfer learning techniques, and the incorporation of contextual information to improve the detection accuracy. This survey provides a comprehensive overview of the current state-of-the-art HS detection methods used in Bengali text and serves as a valuable resource for researchers and practitioners interested in understanding the advancements, challenges, and opportunities in addressing HS in the Bengali language, ultimately assisting in the creation of reliable and effective online platform detection systems.

Abstract Within the Hadoop ecosystem, MapReduce stands as a cornerstone for managing, processing, and mining large-scale datasets. Yet, the absence of efficient solutions for precise estimation of job execution times poses a persistent challenge, impacting task allocation and distribution within Hadoop clusters. In this study, we present a comprehensive machine learning approach for predicting the execution time of MapReduce jobs, encompassing data collection, preprocessing, feature engineering, and model evaluation. Leveraging a rich dataset derived from comprehensive Hadoop MapReduce job traces, we explore the intricate relationship between cluster parameters and job performance. Through a comparative analysis of machine learning models, including linear regression, decision tree, random forest, and gradient-boosted regression trees, we identify the random forest model as the most effective, demonstrating superior predictive accuracy and robustness. Our findings underscore the critical role of features such as data size and resource allocation in determining job performance. With this work, we aim to enhance resource management efficiency and enable more effective utilisation of cloud-based Hadoop clusters for large-scale data processing tasks.

Abstract Synthetic accessibility prediction is a task to estimate how easily a given molecule might be synthesizable in the laboratory, playing a crucial role in computer-aided molecular design. Although synthesis planning programs can determine synthesis routes, their slow processing times make them impractical for large-scale molecule screening. On the other hand, existing rapid synthesis accessibility estimation methods offer speed but typically lack integration with actual synthesis routes and building block information. In this work, we introduce BR-SAScore, an enhanced version of SAScore that integrates the available building block information (B) and reaction knowledge (R) from synthesis planning programs into the scoring process. In particular, we differentiate fragments inherent in building blocks and fragments to be derived from synthesis (reactions) when scoring synthetic accessibility. Compared to existing methods, our experimental findings demonstrate that BR-SAScore offers more accurate and precise identification of a molecule's synthetic accessibility by the synthesis planning program with a fast calculation time. Moreover, we illustrate how BR-SAScore provides chemically interpretable results, aligning with the capability of the synthesis planning program embedded with the same reaction knowledge and available building blocks. Scientific contribution We introduce BR-SAScore, an extension of SAScore, to estimate the synthetic accessibility of molecules by leveraging known building-block and reactivity information. In our experiments, BR-SAScore shows superior prediction performance on predicting molecule synthetic accessibility compared to previous methods, including SAScore and deep-learning models, while requiring significantly less computation time. In addition, we show that BR-SAScore is able to precisely identify the chemical fragment contributing to the synthetic infeasibility, holding great potential for future molecule synthesizability optimization.

Abstract Pretrained Graph Neural Networks have been widely adopted for various molecular property prediction tasks. Despite their ability to encode structural and relational features of molecules, traditional fine-tuning of such pretrained GNNs on the target task can lead to poor generalization. To address this, we explore the adaptation of pretrained GNNs to the target task by jointly training them with multiple auxiliary tasks. This could enable the GNNs to learn both general and task-specific features, which may benefit the target task. However, a major challenge is to determine the relatedness of auxiliary tasks with the target task. To address this, we investigate multiple strategies to measure the relevance of auxiliary tasks and integrate such tasks by adaptively combining task gradients or by learning task weights via bi-level optimization. Additionally, we propose a novel gradient surgery-based approach, Rotation of Conflicting Gradients ( $$\mathop {\texttt{RCGrad}}\limits$$ RCGrad ), that learns to align conflicting auxiliary task gradients through rotation. Our experiments with state-of-the-art pretrained GNNs demonstrate the efficacy of our proposed methods, with improvements of up to 7.7% over fine-tuning. This suggests that incorporating auxiliary tasks along with target task fine-tuning can be an effective way to improve the generalizability of pretrained GNNs for molecular property prediction. Scientific contribution We introduce a novel framework for adapting pretrained GNNs to molecular tasks using auxiliary learning to address the critical issue of negative transfer. Leveraging novel gradient surgery techniques such as $$\mathop {\texttt{RCGrad}}\limits$$ RCGrad , the proposed adaptation framework represents a significant departure from the dominant pretraining fine-tuning approach for molecular GNNs. Our contributions are significant for drug discovery research, especially for tasks with limited data, filling a notable gap in the efficient adaptation of pretrained models for molecular GNNs.

Abstract It is well-accepted that knowledge of a small molecule’s target can accelerate optimization. Although chemogenomic databases are helpful resources for predicting or finding compound interaction partners, they tend to be limited and poorly annotated. Furthermore, unlike genes, compound identifiers are often not standardized, and many synonyms may exist, especially in the biological literature, making batch analysis of compounds difficult. Here, we constructed an open-source annotation and target hypothesis prediction tool that explores some of the largest chemical and biological databases, mining these for both common name, synonyms, and structurally similar molecules. We used this Chemical Analysis and Clustering for Target Identification (CACTI) tool to analyze the Pathogen Box collection, an open-source set of 400 drug-like compounds active against a variety of microbial pathogens. Our analysis resulted in 4,315 new synonyms, 35,963 pieces of new information and target prediction hints for 58 members. Scientific contributions With the employment of this tool, a comprehensive report with known evidence, close analogs and drug-target prediction can be obtained for large-scale chemical libraries that will facilitate their evaluation and future target validation and optimization efforts.

High-Efficiency Video Coding (HEVC) is a video compression standard designed to improve video compression efficiency compared to its predecessor, Advanced Video Coding (AVC). HEVC provides approximately twice the compression efficiency compared to the AVC. In the HEVC system, the most computational complexity module is the Motion Estimation (ME). ME module helps reduce redundancy by compensating for motion during video compression. However, the computational complexity makes it a bottleneck in the design of high-resolution video encoders. This paper proposes an efficient architecture for the Integer ME (IME) module of the HEVC encoder. The proposed architecture introduces an efficient memory usage scheme to support the Full Search motion search algorithm. The full pipeline architecture includes 4096 Processing Elements (PE) and an adder tree to compute the Sum of Absolute Difference (SAD) for every Prediction Unit (PU) partition. The proposed architecture was designed and implemented on Xilinx Virtex-7 XC7VX550T FPGA. Our design achieved up to 275 FPS and approximately 10 FPS at 4 K video for the Search Regions of [Formula: see text] and [Formula: see text] pixels, respectively.

Abstract Detecting spam emails is essential to maintaining the security and integrity of email communication. Existing research has made significant progress in developing effective spam detection models, but challenges remain in improving classification performance and adaptability to evolving spamming techniques. In this study, we propose a novel spam detection model with a comprehensive feature engineering approach that combines term frequency-inverse document frequency (TF-IDF) vectorizer and word embedding features to optimize the feature space. Our contribution lies in integrating semantic-based word embeddings, leveraging pre-existing knowledge to capture the semantic meaning of words and enhance the representation of email texts. To identify the most suitable word embedding technique for our model, we evaluated GloVe, Word2Vec, and FastText. GloVe was selected for its better performance, which is the result of its pre-training on a large and diverse text corpus. Furthermore, the model was evaluated without word embeddings, which did not exhibit the same effectiveness level as our word embedding-based model. Additionally, we utilized the support vector machine as a classifier and hyperparameter tuning technique to identify our model’s most effective parameter values. The proposed model was tested on two datasets. The experimental results showed that our model outperformed the other models discussed in the literature, achieving an accuracy of 99.5% on the SpamAssassin dataset, and 99.28% on the Enron-Spam dataset.

Abstract The present methodology employed in classical control systems is characterized by high costs, significant processing requirements, and inflexibility. In conventional practice, when the controller exhibits instability after being implemented on the hardware, it is often adjusted to achieve stability. However, this approach is not suitable for mass-produced systems like drones, which possess diverse manufacturing tolerances and delicate stability thresholds. The aim of this study is to design and evaluate a controller for a multirotor unmanned aerial vehicle (UAV) system that is capable of adapting its gains in accordance with changes in the system dynamics. The controller utilized in this research employs a Simulink-constructed model that has been taught by reinforcement learning techniques, specifically employing a deep deterministic policy gradient (DDPG) network. The Simulink model of the UAV establishes the framework within which the agent engages in learning through interaction with its surroundings. The DDPG algorithm is an off-policy reinforcement learning technique that operates in continuous action spaces and does not require a model. The efficacy of the cascaded PD controllers and neural network tuner is evaluated. The results revealed that the controller exhibited stability during several flight phases, including take-off, hovering, path tracking, and landing manoeuvres.

Abstract The past few years have seen an emergence of interest in examining the significance of machine learning (ML) in the medical field. Diseases, health emergencies, and medical disorders may now be identified with greater accuracy because of technological advancements and advances in ML. It is essential especially to diagnose individuals with chronic diseases (CD) as early as possible. Our study has focused on analyzing ML’s applicability to predict CD, including cardiovascular disease, diabetes, cancer, liver, and neurological disorders. This study offered a high-level summary of the previous research on ML-based approaches for predicting CD and some instances of their applications. To wrap things up, we compared the results obtained by various studies and the methodologies as well as tools employed by the researchers. The factors or parameters that are responsible for improving the accuracy of the predicting model for different previous works are also identified. For identifying significant features, most of the authors employed a variety of strategies, where least absolute shrinkage and selection (LASSO), minimal-redundancy-maximum-relevance (mRMR), and RELIEF are extensively used methods. It is seen that a wide range of ML approaches, including support vector machine (SVM), random forest (RF), decision tree (DT), naïve Bayes (NB), etc., have been widely used. Also, several deep learning techniques and hybrid models are employed to create CD prediction models, resulting in efficient and reliable clinical decision-making models. For the benefit of the whole healthcare system, we have also offered our suggestions for enhancing the prediction results of CD.

Abstract Time series similarity matrices (informally, recurrence plots or dot-plots), are useful tools for time series data mining. They can be used to guide data exploration, and various useful features can be derived from them and then fed into downstream analytics. However, time series similarity matrices suffer from very poor scalability, taxing both time and memory requirements. In this work, we introduce novel ideas that allow us to scale the largest time series similarity matrices that can be examined by several orders of magnitude. The first idea is a novel algorithm to compute the matrices in a way that removes dependency on the subsequence length. This algorithm is so fast that it allows us to now address datasets where the memory limitations begin to dominate. Our second novel contribution is a multiscale algorithm that computes an approximation of the matrix appropriate for the limitations of the user’s memory/screen-resolution, then performs a local, just-in-time recomputation of any region that the user wishes to zoom-in on. Given that this largely removes time and space barriers, human visual attention then becomes the bottleneck. We further introduce algorithms that search massive matrices with quadrillions of cells and then prioritize regions for later examination by either humans or algorithms. We will demonstrate the utility of our ideas for data exploration, segmentation, and classification in domains as diverse as astronomy, bioinformatics, entomology, and wildlife monitoring.

Abstract While drug combination therapies are of great importance, particularly in cancer treatment, identifying novel synergistic drug combinations has been a challenging venture. Computational methods have emerged in this context as a promising tool for prioritizing drug combinations for further evaluation, though they have presented limited performance, utility, and interpretability. Here, we propose a novel predictive tool, piscesCSM, that leverages graph-based representations to model small molecule chemical structures to accurately predict drug combinations with favourable anticancer synergistic effects against one or multiple cancer cell lines. Leveraging these insights, we developed a general supervised machine learning model to guide the prediction of anticancer synergistic drug combinations in over 30 cell lines. It achieved an area under the receiver operating characteristic curve (AUROC) of up to 0.89 on independent non-redundant blind tests, outperforming state-of-the-art approaches on both large-scale oncology screening data and an independent test set generated by AstraZeneca (with more than a 16% improvement in predictive accuracy). Moreover, by exploring the interpretability of our approach, we found that simple physicochemical properties and graph-based signatures are predictive of chemotherapy synergism. To provide a simple and integrated platform to rapidly screen potential candidate pairs with favourable synergistic anticancer effects, we made piscesCSM freely available online at https://biosig.lab.uq.edu.au/piscescsm/ as a web server and API. We believe that our predictive tool will provide a valuable resource for optimizing and augmenting combinatorial screening libraries to identify effective and safe synergistic anticancer drug combinations. Scientific contribution This work proposes piscesCSM, a machine-learning-based framework that relies on well-established graph-based representations of small molecules to identify and provide better predictive accuracy of syngenetic drug combinations. Our model, piscesCSM, shows that combining physiochemical properties with graph-based signatures can outperform current architectures on classification prediction tasks. Furthermore, implementing our tool as a web server offers a user-friendly platform for researchers to screen for potential synergistic drug combinations with favorable anticancer effects against one or multiple cancer cell lines.

Abstract Purpose Reaction databases are a key resource for a wide variety of applications in computational chemistry and biochemistry, including Computer-aided Synthesis Planning (CASP) and the large-scale analysis of metabolic networks. The full potential of these resources can only be realized if datasets are accurate and complete. Missing co-reactants and co-products, i.e., unbalanced reactions, however, are the rule rather than the exception. The curation and correction of such incomplete entries is thus an urgent need. Methods The SynRBL framework addresses this issue with a dual-strategy: a rule-based method for non-carbon compounds, using atomic symbols and counts for prediction, alongside a Maximum Common Subgraph (MCS)-based technique for carbon compounds, aimed at aligning reactants and products to infer missing entities. Results The rule-based method exceeded 99% accuracy, while MCS-based accuracy varied from 81.19 to 99.33%, depending on reaction properties. Furthermore, an applicability domain and a machine learning scoring function were devised to quantify prediction confidence. The overall efficacy of this framework was delineated through its success rate and accuracy metrics, which spanned from 89.83 to 99.75% and 90.85 to 99.05%, respectively. Conclusion The SynRBL framework offers a novel solution for recalibrating chemical reactions, significantly enhancing reaction completeness. With rigorous validation, it achieved groundbreaking accuracy in reaction rebalancing. This sets the stage for future improvement in particular of atom-atom mapping techniques as well as of downstream tasks such as automated synthesis planning. Scientific Contribution SynRBL features a novel computational approach to correcting unbalanced entries in chemical reaction databases. By combining heuristic rules for inferring non-carbon compounds and common subgraph searches to address carbon unbalance, SynRBL successfully addresses most instances of this problem, which affects the majority of data in most large-scale resources. Compared to alternative solutions, SynRBL achieves a dramatic increase in both success rate and accurary, and provides the first freely available open source solution for this problem.

Abstract Motivation Retrosynthesis planning poses a formidable challenge in the organic chemical industry, particularly in pharmaceuticals. Single-step retrosynthesis prediction, a crucial step in the planning process, has witnessed a surge in interest in recent years due to advancements in AI for science. Various deep learning-based methods have been proposed for this task in recent years, incorporating diverse levels of additional chemical knowledge dependency. Results This paper introduces UAlign, a template-free graph-to-sequence pipeline for retrosynthesis prediction. By combining graph neural networks and Transformers, our method can more effectively leverage the inherent graph structure of molecules. Based on the fact that the majority of molecule structures remain unchanged during a chemical reaction, we propose a simple yet effective SMILES alignment technique to facilitate the reuse of unchanged structures for reactant generation. Extensive experiments show that our method substantially outperforms state-of-the-art template-free and semi-template-based approaches. Importantly, our template-free method achieves effectiveness comparable to, or even surpasses, established powerful template-based methods. Scientific contribution We present a novel graph-to-sequence template-free retrosynthesis prediction pipeline that overcomes the limitations of Transformer-based methods in molecular representation learning and insufficient utilization of chemical information. We propose an unsupervised learning mechanism for establishing product-atom correspondence with reactant SMILES tokens, achieving even better results than supervised SMILES alignment methods. Extensive experiments demonstrate that UAlign significantly outperforms state-of-the-art template-free methods and rivals or surpasses template-based approaches, with up to 5% (top-5) and 5.4% (top-10) increased accuracy over the strongest baseline.

Non-Collusive Interest Flooding Attacks (NCIFA) disrupt Named Data Networks' (NDNs) seamless communication and content distribution through QoS degradation. This work proposes 2TierEdge-Defense: a framework based on Long Short Term Memory (LSTM) to detect NCIFA in NDN at the edge content routers. The framework has been evaluated on large-scale Rocketfuel topologies for AT&T Internet Service Provider. The 2TierEdge-Defense framework consists of attack detection and mitigation modules with detection at edge router and their interfaces. During offline training and evaluation, the 2TierEdge-Defense framework can detect NCIFA with 0.92 (F1-score) and 0.84 (F1-score) at edge routers and their interfaces, respectively, with cascaded F1-score of 0.9872. Upon detection, the mitigation strategy in the 2TierEdge-Defense framework can improve the QoS metrics as quickly as in 0.45 seconds with an overall F1 score as high as 0.9872. The 2TierEdge-Defense is evaluated for scalability topology scenarios to ensure better performance when deployed on NDN content routers with a cascaded miss rate of 0.1 and a false alarm rate of 0.07.

Abstract In the operation and control of power systems, load frequency control (LFC) plays a critical role in ensuring the stability and reliability of interconnected power systems. Modern power systems with significant penetration of highly variable and intermittent renewable sources present new challenges that make traditional control strategies ineffective. To address these new challenges, this paper proposes a novel LFC strategy that employs a cascaded fractional-order proportional integral-fractional-order proportional integral derivative with a derivative filter (FOPI-FOPIDN) as a controller. The parameters of the FOPI-FOPIDN are optimised using a variant of the particle swarm optimization (PSO) in the literature called ADIWACO. The effectiveness and scalability of the proposed strategy are validated by extensive simulations conducted on two- and three-area test systems and performance comparisons with recent LFC control strategies in the literature. The performance metrics used for the evaluation are ITAE values, deviations in the power flows in the tie-lines, and deviations in the frequencies of the control areas with the power systems subjected to diverse load and RES generation disturbances in several experimental scenarios. Governor dead band, communication time delay, and generation rate constraints are considered in one of the scenarios for more realistic evaluation. Again, the controller’s robustness to uncertain model parameters is validated by varying the parameters of the three-area test system by ± 50%. The simulation results obtained confirm the controller’s robustness and its superiority over the comparison LFC strategies in terms of the above performance metrics.

Abstract ALS (Amyotrophic Lateral Sclerosis) is a fatal neurodegenerative disease of the human motor system. It is a group of progressive diseases that affects the nerve cells in the brain and spinal cord that control the muscle movement of the body hence, detection and classification of ALS at the right time is considered to be one of the vital aspects that can save the life of humans. Therefore, in various studies, different AI techniques are used for the detection of ALS, however, these methods are considered to be ineffectual in terms of identifying the disease due to the employment of ineffective algorithms. Hence, the proposed model utilizes Modified Principal Component Analysis (MPCA) and Modified Random Forest (MRF) for performing dimensionality reduction of all the potential features considered for effective classification of the ALS presence and absence of ALS causing mutation in the corresponding gene. The MPCA is adapted for capturing all the Low-Importance Data transformation. Furthermore, The MPCA is objected to performing three various approaches: Covariance Matrix Correlation, Eigen Vector- Eigenvalue decomposition, and selecting the desired principal components. This is done in aspects of implying the LI (Lower-Importance) Data Transformation. By choosing these potential components without any loss of features ensures better viability of selecting the attributes for ALS-causing gene classification. This is followed by the classification of the proposed model by using Modified RF by updating the clump detector technique. The clump detector is proceeded by clustering approach using K-means, and the data reduced by their dimension are grouped accordingly. These clustered data are analyzed either for ALS causing or devoid of causing ALS. Finally, the model’s performance is assessed using different evaluation metrics like accuracy, recall, F1 score, and precision, and the proposed model is further compared with the existing models to assess the efficacy of the proposed model.

Abstract Nowadays, due to the constantly growing amount of textual information, automatic text summarization constitutes an important research area in natural language processing. In this work, we present a novel framework that combines semantic graph representations along with deep learning predictions to generate abstractive summaries of single documents, in an effort to utilize a semantic representation of the unstructured textual content in a machine-readable, structured, and concise manner. The overall framework is based on a well defined methodology for performing semantic graph parsing, graph construction, graph transformations for machine learning models and deep learning predictions. The employed semantic graph representation focuses on using the model of abstract meaning representation. Several combinations of graph construction and graph transformation methods are investigated to specify the most efficient of them for the machine learning models. Additionally, a range of deep learning architectures is examined, including a sequence-to-sequence attentive network, reinforcement learning, transformer-based architectures, and pre-trained neural language models. In this direction, a semantic graph representation of an original text is extracted, and then the present framework formulates the problem as a graph-to-summary learning problem to predict a summary of an original text. To the best of our knowledge, this formulation of graph-to-summary predictions in abstractive text summarization, without other intermediate steps in the machine learning phase, has not been presented in the relevant literature. Another important contribution is an introduction of a measure for assessing the factual consistency of the generated summaries in an effort to provide a qualitative evaluation. To assess the framework, an extensive experimental procedure is presented that uses popular datasets to evaluate key aspects of the proposed approach. The obtained results exhibit promising performance, validating the robustness of the proposed framework.