Search

In the recent years, the academic paper influence analysis has been widely studied due to its potential applications in the multiple areas of science information metric and retrieval. By identifying the academic influence of papers, authors, etc., we can directly support researchers to easily reach academic papers. These recommended candidate papers are not only highly relevant with their desired research topics but also highly-attended by the research community within these topics. For very recent years, the rapid developments of academic networks, like Google Scholar, Research Gate, CiteSeerX, etc., have significantly boosted the number of new published papers annually. It also helps to strengthen the borderless cooperation between researchers who are interested on the same research topics. However, these current academic networks still lack the capabilities of provisioning researchers deeper into most-influenced papers. They also largely ignore quite/irrelevant papers, which are not fully related with their current interest topics. Moreover, the distributions of topics within these academic papers are considered as varying and it is difficult to extract the main concentrated topics in these papers. Thus, it leads to challenges for researchers to find their appropriated/high-qualified reference resources while doing researches. To overcome this limitation, in this paper, we proposed a novel approach of paper influence analysis through their content-based and citation relationship-based analyses within the biographical network. In order to effectively extract the topic-based relevance from papers, we apply the integrated graph-based citation relationship analysis with topic modeling approach to automatically learn the distributions of keyword-based labeled topics in forms of unsupervised learning approach, named as TopCite. Then, we base on the constructed graph-based paper–topic structure to identify their relevancy levels. Upon the identified relevancy levels between papers, we can support for improving the accuracy performance of other bibliographic network mining tasks, such as paper similarity measurement, recommendation, etc. Extensive experiments in real-world AMiner bibliographic dataset demonstrate the effectiveness of our proposed ideas in this paper. [ABSTRACT FROM AUTHOR]

To address the issue of insufficient testing caused by the continuous reduction of software development cycles, many organizations maintain bug repositories and bug tracking systems to ensure real-time updates of bugs. However, each day, a large number of bugs is discovered and sent to the repository, which imposes a heavy workload on bug fixers. Therefore, effective bug deduplication and triage are of great significance in software development. This paper provides a comprehensive investigation and survey of the recent developments in bug deduplication and triage. The study begins by outlining the roadmap of the existing literature, including the research trends, mathematical models, methods, and commonly used datasets in recent years. Subsequently, the paper summarizes the general process of the methods from two perspectives—runtime information-based and bug report-based perspectives—and provides a detailed overview of the methodologies employed in relevant works. Finally, this paper presents a detailed comparison of the experimental results of various works in terms of usage methods, datasets, accuracy, recall rate, and F1 score. Drawing on key findings, such as the need to improve the accuracy of runtime information collection and refine the description information in bug reports, we propose several potential future research directions in the field, such as stack trace enrichment and the combination of new NLP models. [ABSTRACT FROM AUTHOR]

Much of the knowledge and information needed for enabling high-quality clinical research is stored in free-text format. Natural language processing (NLP) has been used to extract information from these sources at scale for several decades. This paper aims to present a comprehensive review of clinical NLP for the past 15 years in the UK to identify the community, depict its evolution, analyse methodologies and applications, and identify the main barriers. We collect a dataset of clinical NLP projects (n = 94; £ = 41.97 m) funded by UK funders or the European Union's funding programmes. Additionally, we extract details on 9 funders, 137 organisations, 139 persons and 431 research papers. Networks are created from timestamped data interlinking all entities, and network analysis is subsequently applied to generate insights. 431 publications are identified as part of a literature review, of which 107 are eligible for final analysis. Results show, not surprisingly, clinical NLP in the UK has increased substantially in the last 15 years: the total budget in the period of 2019–2022 was 80 times that of 2007–2010. However, the effort is required to deepen areas such as disease (sub-)phenotyping and broaden application domains. There is also a need to improve links between academia and industry and enable deployments in real-world settings for the realisation of clinical NLP's great potential in care delivery. The major barriers include research and development access to hospital data, lack of capable computational resources in the right places, the scarcity of labelled data and barriers to sharing of pretrained models. [ABSTRACT FROM AUTHOR]

In the biomedical field due to tremendous medical research, every year the medical papers generated is very huge. In order to process these tremendous amounts of textual data, the information retrieval techniques are combined with machine learning methods like natural language processing (NLP) for finding the useful insights in data for NLP tasks. In this work, we have addressed the challenges involved in existing NLP techniques like Word2Vec, Doc2Vec, and Biosent2Vec and followed by the traditional classification of articles with Concept2Vec and random forest classification. Concept vector techniques are emerging methods in NLP and in information retrieval systems. Pretrained concept encoders are available in Machine learning, whereas no such exists in biomedical domain. In this paper a proposal on Concept2Vec trained from biomedical text is highlighted. Our work accurately emphasizes on the quality of semantics defined between the medical terms, which can help the medical practitioners in taking errorless decisions. In this work 20 thousand biomedical documents are considered for Concept2Vec embeddings and the proposed approach have shown best concept similarity in terms of relevant measures compared with existing approaches such as Word2Vec, global vectors (GloVe), fastText, and BioSentVec, bidirectional encoder representations from transformers (BERT), biomedical-BERT. [ABSTRACT FROM AUTHOR]

Text classification is an essential task for many applications related to the Natural Language Processing domain. It can be applied in many fields, such as Information Retrieval, Knowledge Extraction, and Knowledge modeling. Even though the importance of this task, Arabic Text Classification tools still suffer from many problems and remain incapable of responding to the increasing volume of Arabic content that circulates on the web or resides in large databases. This paper introduces a novel machine learning-based approach that exclusively uses hybrid (stylistic and semantic) features. First, we clean the Arabic documents and translate them to English using translation tools. Consequently, the semantic features are automatically extracted from the translated documents using an existing database of English topics. Besides, the model automatically extracts from the textual content a set of stylistic features such as word and character frequencies and punctuation. Therefore, we obtain 3 types of features: semantic, stylistic and hybrid. Using each time, a different type of feature, we performed an in-depth comparison study of nine well-known Machine Learning models to evaluate our approach and used a standard Arabic corpus. The obtained results show that Neural Network outperforms other models and provides good performances using hybrid features (F1-score = 0.88%). [ABSTRACT FROM AUTHOR]

Biomedical-named entity recognition (bNER) is critical in biomedical informatics. It identifies biomedical entities with special meanings, such as people, places, and organizations, as predefined semantic types in electronic health records (EHR). bNER is essential for discovering novel knowledge using computational methods and Information Technology. Early bNER systems were configured manually to include domain-specific features and rules. However, these systems were limited in handling the complexity of the biomedical text. Recent advances in deep learning (DL) have led to the development of more powerful bNER systems. DL-based bNER systems can learn the patterns of biomedical text automatically, making them more robust and efficient than traditional rule-based systems. This paper reviews the healthcare domain of bNER, using DL techniques and artificial intelligence in clinical records, for mining treatment prediction. bNER-based tools are categorized systematically and represent the distribution of input, context, and tag (encoder/decoder). Furthermore, to create a labeled dataset for our machine learning sentiment analyzer to analyze the sentiment of a set of tweets, we used a manual coding approach and the multi-task learning method to bias the training signals with domain knowledge inductively. To conclude, we discuss the challenges facing bNER systems and future directions in the healthcare field. [ABSTRACT FROM AUTHOR]

Zero-Shot Learning (ZSL) aims to classify images of new categories in the testing phase without labeled images during training, using examples from categories with labeled images and some auxiliary information. The auxiliary information includes semantic attributes, textual descriptions, word embeddings, etc., for both labeled and unlabeled classes, utilizing Natural Language Processing (NLP) approaches. The word embeddings created are limited by the semantic attributes and textual descriptions where the semantics of categories are insufficient. In this paper, introduces a study for Generalized Zero-Shot Learning (GZSL), a type of ZSL, by integrating the rich semantics offered by ontology. Semantic attributes used for semantic representation are supported by ontology. Variational Autoencoder (VAE) and Generative Adversarial Network (GAN) network architectures are used together to synthesize visual features. Our work was evaluated on the AWA2 dataset, and improvement in GZSL performance was achieved. [ABSTRACT FROM AUTHOR]

Coreference Resolution is an essential task for Natural Language Processing (NLP) application, which has a paramount impact on the performance of text summarization, machine translation, text classification, and recognizing textual entailment. Mention Detection (MD) is the core component of the coreference resolution task and is additionally a process of extraction of all possible mentions from the text. Mention is referred to as a textual representation of entities in the text, such as Name, Nominal, and Pronominal mentions. The mentions appear in the text using different representations but indicating the same entity. The performance of an MD module positively affects the performance of NLP tasks such as Coreference resolution, Relation Extraction, Information retrieval, Information extraction, etc. Incorrect identification of mentions in the text severely affects the efficiency of the coreference resolution task. This paper aims to provide a comprehensive overview for the state of the art of mention detection approaches, which is utilized in the coreference resolution task and explains the importance of MD in Coreference resolution. The subsisting approaches are classified based on the underlying techniques adopted by each approach in three categories: Rule-based mention detection, Statistics-based mention detection, and Deep learning-based mention detection. The performance of deep learning is improving as more data and more powerful computing resources become available. This study endeavors to provide a comparative analysis of various mention detection approaches and help the researchers to assimilate knowledge about the mention detection approaches from sundry aspects. [ABSTRACT FROM AUTHOR]

The internet is surrounded by uncertain information which necessitates the usage of natural language processing and soft computing techniques to extract the relevant documents. The relevant results are retrieved using the query expansion technique which is mainly formulated using the machine learning or deep learning concepts in the existing literature. This paper presents a hybrid group mean-based optimizer-enhanced chimp optimization (GMBO-ECO) algorithm for pseudo-relevance-based query expansion, whereby the actual queries are expanded with their related keywords. The hybrid GMBO-ECO algorithm mainly expands the query based on the terms that have a strong interrelationship with the actual query. To generate the word embeddings, a Word2Vec paradigm is used which learns the word association from large text corpora. The useful context in the text is identified using the improved iterative deep learning framework which determines the user's intent for the current web search. This step reduces the mismatch of the words and improves the performance of query retrieval. The weak terms are eliminated and the candidate query terms for optimal query expansion are improved via an Okapi measure and cosine similarity techniques. The proposed methodology has been compared to the state-of-the-art methods with and without a query expansion approach. Moreover, the proposed optimal query expansion technique has shown a substantial improvement in terms of a normalized discounted cumulative gain of 0.87, a mean average precision of 0.35, and a mean reciprocal rank of 0.95. The experimental results show the efficiency of the proposed methodology in retrieving the appropriate response for information retrieval. The most common applications for the proposed method are search engines. [ABSTRACT FROM AUTHOR]

This paper proposes an interactive dialog system, called AidIR, to aid information retrieval. AidIR allows users to retrieve information on diseases resulting from coronaviruses and diseases transmitted by vector mosquitoes with natural language interaction and Line chat media. In a subjective evaluation, we asked 20 users to rate the intuitiveness, usability, and user experience of AidIR with a range between −2 and 2. Moreover, we also asked these users to answer yes–no questions to evaluate AidIR and provide feedback. The average scores of intuitiveness, usability, and user experience are 0.8, 0.8, and 1.05, respectively. The yes–no questions demonstrated that AidIR is better than systems using the graphical user interface in mobile phones and single-turn dialog systems. According to user feedback, AidIR is more convenient for information retrieval. Moreover, we designed a new loss function to jointly train a BERT model for domain classification and sequence label tasks. The accuracy of both tasks is 92%. Finally, we trained the dialog policy network with supervised learning tasks and deployed the reinforcement learning algorithm to allow AidIR to continue learning the dialog policy. [ABSTRACT FROM AUTHOR]

In the era of rapid paper publications in various venues, automatic citation recommendations would be highly useful to researchers when they write papers. Local citation recommendation aims to recommend possible papers to cite given local citation contexts. Previous work mainly computes the similarity score between citation contexts and cited papers on a one-to-one basis, which is quite time-consuming. We train a pair of neural network encoders that map citation contexts and all possible cited papers to the same vector space, respectively. After that, we index the positions of all cited papers in the vector space. This makes our process for searching recommended papers considerably faster. On the other hand, existing methods tend to recommend papers that are highly similar to each other, which makes recommendations lack diversity. Therefore, we extend our algorithm to perform multi-topic recommendations. We generate multi-topic training examples based on the index we mentioned earlier. Furthermore, we specially design a multi-group contrastive learning method to train our model so that it can distinguish different topics. Empirical experiments show that our model outperforms previous methods by a wide margin. Our model is also light weighted and has been deployed online so that researchers can use it to obtain recommended citations for their own paper in real-time. • Proposed FLCR algorithm with sentence-transformer & k-d tree. • Introduced multi-topic citation recommendation for diverse contexts. • Developed large-scale dataset with 1.7 million citation contexts for evaluation. • Demonstrated significant performance improvement over previous methods. • Deployed demo for real-time citation recommendations. [ABSTRACT FROM AUTHOR]

Summary: The frequent user interactions happening in the form of textual contents like reviews, ratings, tags, blogs, testimonials, and so forth transformed the social media platform into a contextualized and personalized data warehouse focusing its users' unique likes and dislikes. The huge volume of social media content makes it difficult for the end users to consume relevant information by themselves. The need of a tool to deal with such scenario leads to the development of recommendation systems. This work proposes an ensemble multi‐stage recommender system with sentiment based clustering to deal with social media text corpus where each stage performing unique functionalities of information retrieval, natural language processing, user segmentation, prediction, and recommendation generation. The proposed system leverages a hybrid approach of content‐based, collaborative, and demographic filtering techniques to predict and recommend contents, products, or services according to user interests. The experimental results gathered using standard datasets are promising and found more efficient than the traditional approaches. [ABSTRACT FROM AUTHOR]

Sentiment analysis is a practical technique that allows businesses, researchers, governments, politicians, and organizations to know about people's sentiments, which play an important role in decision-making processes. Sentiment classification techniques are mainly divided into lexicon-based methods, machine learning methods, and hybrid methods. There are limitations in each approach; Traditional machine learning approaches are based on complex features extraction process, and lexicon-based approaches suffer from scalability and are limited by unreliable sentiment lexicons that are commonly created manually by experts. In this paper, we seek to improve the performance of machine learning techniques by integrating it with our enhanced lexicon based technique Sum-of-Votes Model. Sum-of-Votes model is a generic extendable lexicon based model that beats traditional lexicon based models in accuracy and provides good solutions to previous challenges and drawbacks such as scalability, domain dependency, and unreliability, but its accuracy was 81.62%. So, in this paper we proposed a novel framework based upon both Sum-of-Votes and Bag-of-Words models; we applied them, then their outputs were fed as features to Machine Learning Classifiers. We got higher accuracy than all the individual lexicons and the entire old framework. [ABSTRACT FROM AUTHOR]

This paper introduces a new enhanced Arabic stemming algorithm for solving the information retrieval problem, especially in medical documents. Our proposed algorithm is a light stemming algorithm for extracting stems and roots from the input data. One of the main challenges facing the light stemming algorithm is cutting off the input word, to extract the initial segments. When initiating the light stemmer with strong initial segments, the final extracting stems and roots will be more accurate. Therefore, a new enhanced segmentation based on deploying the Direct Acyclic Graph (DAG) model is utilized. In addition to extracting the powerful initial segments, the main two procedures (i.e., stems and roots extraction), should be also reinforced with more efficient operators to improve the final outputs. To validate the proposed enhanced stemmer, four data sets are used. The achieved stems and roots resulted from our proposed light stemmer are compared with the results obtained from five other well-known Arabic light stemmers using the same data sets. This evaluation process proved that the proposed enhanced stemmer outperformed other comparative stemmers. [ABSTRACT FROM AUTHOR]

Copyright of Procesamiento del Lenguaje Natural is the property of Sociedad Espanola para el Procesamiento del Lenguaje Natural and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Language models (LMs) such as BERT and GPT have revolutionized natural language processing (NLP). However, the medical field faces challenges in training LMs due to limited data access and privacy constraints imposed by regulations like the Health Insurance Portability and Accountability Act (HIPPA) and the General Data Protection Regulation (GDPR). Federated learning (FL) offers a decentralized solution that enables collaborative learning while ensuring data privacy. In this study, we evaluated FL on 2 biomedical NLP tasks encompassing 8 corpora using 6 LMs. Our results show that: (1) FL models consistently outperformed models trained on individual clients' data and sometimes performed comparably with models trained with polled data; (2) with the fixed number of total data, FL models training with more clients produced inferior performance but pre-trained transformer-based models exhibited great resilience. (3) FL models significantly outperformed pre-trained LLMs with few-shot prompting. [ABSTRACT FROM AUTHOR]

Clinical narratives commonly use acronyms without explicitly defining their long forms. This makes it difficult to automatically interpret their sense as acronyms tend to be highly ambiguous. Supervised learning approaches to their disambiguation in the clinical domain are hindered by issues associated with patient privacy and manual annotation, which limit the size and diversity of training data. In this study, we demonstrate how scientific abstracts can be utilised to overcome these issues by creating a large automatically annotated dataset of artificially simulated global acronyms. A neural network trained on such a dataset achieved the F1-score of 95% on disambiguation of acronym mentions in scientific abstracts. This network was integrated with multiword term recognition to extract a sense inventory of acronyms from a corpus of clinical narratives on the fly. Acronym sense extraction achieved the F1-score of 74% on a corpus of radiology reports. In clinical practice, the suggested approach can be used to facilitate development of institution-specific inventories. [ABSTRACT FROM AUTHOR]

OBJECTIVE: To compare the accuracy of an artificial intelligence chatbot to clinical practice guidelines (CPGs) recommendations for providing answers to complex clinical questions on lumbosacral radicular pain. DESIGN: Cross-sectional study. METHODS: We extracted recommendations from recent CPGs for diagnosing and treating lumbosacral radicular pain. Relative clinical questions were developed and queried to OpenAI's ChatGPT (GPT-3.5). We compared ChatGPT answers to CPGs recommendations by assessing the (1) internal consistency of ChatGPT answers by measuring the percentage of text wording similarity when a clinical question was posed 3 times, (2) reliability between 2 independent reviewers in grading ChatGPT answers, and (3) accuracy of ChatGPT answers compared to CPGs recommendations. Reliability was estimated using Fleiss' kappa (k) coefficients, and accuracy by interobserver agreement as the frequency of the agreements among all judgments. RESULTS: We tested 9 clinical questions. The internal consistency of text ChatGPT answers was unacceptable across all 3 trials in all clinical questions (mean percentage of 49%, standard deviation of 15). Intrareliability (reviewer 1: k = 0.90, standard error [SE] = 0.09; reviewer 2: k = 0.90, SE = 0.10) and interreliability (k = 0.85, SE = 0.15) between the 2 reviewers was "almost perfect." Accuracy between ChatGPT answers and CPGs recommendations was slight, demonstrating agreement in 33% of recommendations. CONCLUSION: ChatGPT performed poorly in internal consistency and accuracy of the indications generated compared to clinical practice guideline recommendations for lumbosacral radicular pain. [ABSTRACT FROM AUTHOR]

Cross-linked threat, vulnerability, and defensive mitigation knowledge is critical in defending against diverse and dynamic cyber threats. Cyber analysts consult it by deductively or inductively creating a chain of reasoning to identify a threat starting from indicators they observe or vice versa. Cyber hunters use it abductively to reason when hypothesizing specific threats. Threat modelers use it to explore threat postures. We aggregate five public sources of threat knowledge and three public sources of knowledge that describe cyber defensive mitigations, analytics, and engagements and which share some unidirectional links between them. We unify the sources into a graph, and in the graph, we make all unidirectional cross-source links bidirectional. This enhancement of the knowledge makes the questions that analysts and automated systems formulate easier to answer. We demonstrate this in the context of various cyber analytic and hunting tasks as well as modeling and simulations. Because the number of linked entries is very sparse, to further increase the analytic utility of the data, we use natural language processing and supervised machine learning to identify new links. These two contributions demonstrably increase the value of the knowledge sources for cyber security activities. [ABSTRACT FROM AUTHOR]

With the rapid development of the Internet and information technology, people are able to create multimedia data such as pictures or videos anytime and anywhere. Efficient multimedia processing tools are needed for the vast video data. The video moment localization task aims to locate the video moment which best matches the query in the untrimmed video. Existing text-guided methods only consider single-scale text features, which cannot fully represent the semantic features of text, and also do not consider the masking of crucial information in the video by text information when using text to guide the extraction of video features. To solve the above problems, we propose a video moment localization network based on text multi-semantic clues guidance. Specifically, we first design a text encoder based on fusion gate to better capture the semantic information in the text through multi-semantic clues composed of word embedding, local features and global features. Then text guidance module guides the extraction of video features by text semantic features to highlight the video features related to text semantics. Experimental results on two datasets, Charades-STA and ActivityNet Captions, show that our approach provides significant improvements over state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Rising incidence and mortality of cancer have led to an incremental amount of research in the field. To learn from preexisting data, it has become important to capture maximum information related to disease type, stage, treatment, and outcomes. Medical imaging reports are rich in this kind of information but are only present as free text. The extraction of information from such unstructured text reports is labor-intensive. The use of Natural Language Processing (NLP) tools to extract information from radiology reports can make it less time-consuming as well as more effective. In this study, we have developed and compared different models for the classification of lung carcinoma reports using clinical concepts. This study was approved by the institutional ethics committee as a retrospective study with a waiver of informed consent. A clinical concept-based classification pipeline for lung carcinoma radiology reports was developed using rule-based as well as machine learning models and compared. The machine learning models used were XGBoost and two more deep learning model architectures with bidirectional long short-term neural networks. A corpus consisting of 1700 radiology reports including computed tomography (CT) and positron emission tomography/computed tomography (PET/CT) reports were used for development and testing. Five hundred one radiology reports from MIMIC-III Clinical Database version 1.4 was used for external validation. The pipeline achieved an overall F1 score of 0.94 on the internal set and 0.74 on external validation with the rule-based algorithm using expert input giving the best performance. Among the machine learning models, the Bi-LSTM_dropout model performed better than the ML model using XGBoost and the Bi-LSTM_simple model on internal set, whereas on external validation, the Bi-LSTM_simple model performed relatively better than other 2. This pipeline can be used for clinical concept-based classification of radiology reports related to lung carcinoma from a huge corpus and also for automated annotation of these reports. [ABSTRACT FROM AUTHOR]

Automatic Question Answering (QA) has been successfully applied in various domains such as search engines and chatbots. Biomedical QA (BQA), as an emerging QA task, enables innovative applications to effectively perceive, access, and understand complex biomedical knowledge. There have been tremendous developments of BQA in the past two decades, which we classify into five distinctive approaches: classic, information retrieval, machine reading comprehension, knowledge base, and question entailment approaches. In this survey, we introduce available datasets and representative methods of each BQA approach in detail. Despite the developments, BQA systems are still immature and rarely used in real-life settings. We identify and characterize several key challenges in BQA that might lead to this issue, and we discuss some potential future directions to explore. [ABSTRACT FROM AUTHOR]

Toponym recognition, or the challenge of detecting place names that have a similar referent, is involved in a number of activities connected to geographical information retrieval and geographical information sciences. This research focuses on recognizing Chinese toponyms from social media communications. While broad named entity recognition methods are frequently used to locate places, their accuracy is hampered by the many linguistic abnormalities seen in social media posts, such as informal sentence constructions, name abbreviations, and misspellings. In this study, we describe a Chinese toponym identification model based on a hybrid neural network that was created with these linguistic inconsistencies in mind. Our method adds a number of improvements to a standard bidirectional recurrent neural network model to help with location detection in social media messages. We demonstrate the results of a wide-ranging evaluation of the performance of different supervised machine learning methods, which have the natural advantage of avoiding human design features. A set of controlled experiments with four test datasets (one constructed and three public datasets) demonstrates the performance of supervised machine learning that can achieve good results on the task, significantly outperforming seven baseline models. [ABSTRACT FROM AUTHOR]

Texts have become an important spatial data resource. Interpretation of unstructured geoscience texts using natural language processing methods can effectively facilitate the discovery and retrieval of geographic information. Yet studies on the extraction of spatial information from textual geoscience data are limited compared to digital geoscience data. In this work, a machine learning approach is proposed for mining spatial relations in Chinese geological texts. The approach views spatial relation extraction as a sequence labeling problem, avoids the division of relation categories, and enables mining fine-grained spatial relations. The extracted geological texts commonly describe three-dimensional spatial relations among regions, strata, and lithologies. The extracted spatial relations are classified into three major categories (topological relations, absolute directional relations and relative directional relations) and 14 subcategories. We validated the proposed model with a test dataset, constructed visual displays of the extracted spatial relations on different topics, and quantified the uncertainty in the process from spatial entity recognition to spatial relation extraction. With the detailed portrayal of these spatial relations, this study provides support for solving theoretical and practical problems of cognition, prediction, decision-making, and evaluation in geoscience. [ABSTRACT FROM AUTHOR]

Optical character recognition (OCR) is one of the most popular techniques used for converting printed documents into machine-readable ones. While OCR engines can do well with modern text, their performance is unfortunately significantly reduced on historical materials. Additionally, many texts have already been processed by various out-of-date digitisation techniques. As a consequence, digitised texts are noisy and need to be post-corrected. This article clarifies the importance of enhancing quality of OCR results by studying their effects on information retrieval and natural language processing applications. We then define the post-OCR processing problem, illustrate its typical pipeline, and review the state-of-the-art post-OCR processing approaches. Evaluation metrics, accessible datasets, language resources, and useful toolkits are also reported. Furthermore, the work identifies the current trend and outlines some research directions of this field. [ABSTRACT FROM AUTHOR]

Information retrieval IR is the process of finding information (generally documents) that matches the needs of the user. One way to improve the search effectiveness, as well as the quality of text categorization is to build an effective stemmer that helps to match users’ queries with relevant documents in IR and reduce the space of textual representation in TC. This has been always an interesting research topic in IR and TC. We can define stemming as the process of reducing inflected and derived words to their reduced forms (stems or roots). Many stemmers have been developed for different languages, but there is always many weaknesses and problems. In the present work, we have developed a multilingual stemming approach, based on the extraction of the word root and that exploits the technique of n-grams of characters. Our experiments have been done on three languages which are: Arabic, English, and French. [ABSTRACT FROM AUTHOR]

Traditional clinical named entity recognition methods fail to balance the effectiveness of feature extraction of unstructured text and the complexity of neural network models. We propose a model based on ALBERT and a multihead attention (MHA) mechanism to solve this problem. Structurally, the model first obtains character-level word embeddings through the ALBERT pretraining language model, then inputs the word embeddings into the iterated dilated convolutional neural network model to quickly extract global semantic information, and decodes the predicted labels through conditional random fields to obtain the optimal label sequence. Also, we apply the MHA mechanism to capture intercharacter dependencies from multiple aspects. Furthermore, we use the RAdam optimizer to boost the convergence speed and improve the generalization ability of our model. Experimental results show that our model achieves an F1 score of 85.63% on the CCKS-2019 dataset—an increase of 4.36% compared to the baseline model. [ABSTRACT FROM AUTHOR]

Reviewing the metabolomics literature is becoming increasingly difficult because of the rapid expansion of relevant journal literature. Text-mining technologies are therefore needed to facilitate more efficient literature reviews. Here we contribute a standardised corpus of full-text publications from metabolomics studies and describe the development of two metabolite named entity recognition (NER) methods. These methods are based on Bidirectional Long Short-Term Memory (BiLSTM) networks and each incorporate different transfer learning techniques (for tokenisation and word embedding). Our first model (MetaboListem) follows prior methodology using GloVe word embeddings. Our second model exploits BERT and BioBERT for embedding and is named TABoLiSTM (Transformer-Affixed BiLSTM). The methods are trained on a novel corpus annotated using rule-based methods, and evaluated on manually annotated metabolomics articles. MetaboListem (F1-score 0.890, precision 0.892, recall 0.888) and TABoLiSTM (BioBERT version: F1-score 0.909, precision 0.926, recall 0.893) have achieved state-of-the-art performance on metabolite NER. A training corpus with full-text sentences from >1000 full-text Open Access metabolomics publications with 105,335 annotated metabolites was created, as well as a manually annotated test corpus (19,138 annotations). This work demonstrates that deep learning algorithms are capable of identifying metabolite names accurately and efficiently in text. The proposed corpus and NER algorithms can be used for metabolomics text-mining tasks such as information retrieval, document classification and literature-based discovery and are available from the omicsNLP GitHub repository. [ABSTRACT FROM AUTHOR]

Synthetic speech's applications sector is rapidly growing, as well as the quality of TTS systems is continuously improving. Speech synthesis solutions are becoming relatively cheap for average users, making them more appropriate for regular usage, especially as NLP, Machine Learning, and Artificial Intelligence continue to develop where Synthetic speech may also be utilised in a variety of industries, and with effective integration, money can indeed be saved. [ABSTRACT FROM AUTHOR]

As resources for conservation are limited, gathering and analysing information from digital platforms can help investigate the global biodiversity crisis in a cost‐efficient manner. Development and application of methods for automated content analysis of digital data sources are especially important in the context of investigating human–nature interactions.In this study, we introduce novel application methods to automatically collect and analyse textual data on species of conservation concern from digital platforms. An end‐to‐end pipeline is constructed that begins from searching and downloading news articles about species listed in Appendix I of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) along with news articles from specific Twitter handles and proceeds with implementing natural language processing and machine learning methods to filter and retain only relevant articles. A crucial aspect here is the automatic annotation of training data, which can be challenging in many machine learning applications. A Named Entity Recognition model is then used to extract additional relevant information for each article.The data collected over a 1‐month period included 15,088 articles focusing on 585 species listed in Appendix I of CITES. The accuracy of the neural network to detect relevant articles was 95.91% while the Named Entity recognition model helped extract information on prices, location and quantities of traded animals and plants. A regularly updated database, which can be queried and analysed for various research purposes and to inform conservation decision making, is generated by the system.The results demonstrate that natural language processing can be used successfully to extract information from digital text content. The proposed methods can be applied to multiple digital data platforms at the same time and used to investigate human–nature interactions in conservation science and practice. [ABSTRACT FROM AUTHOR]

Contractors responsible for the whole execution of engineering, procurement, and construction (EPC) projects are exposed to multiple risks due to various unbalanced contracting methods such as lump-sum turn-key and low-bid selection. Although systematic risk management approaches are required to prevent unexpected damage to the EPC contractors in practice, there were no comprehensive digital toolboxes for identifying and managing risk provisions for ITB and contract documents. This study describes two core modules, Critical Risk Check (CRC) and Term Frequency Analysis (TFA), developed as a digital EPC contract risk analysis tool for contractors, using artificial intelligence and text-mining techniques. The CRC module automatically extracts risk-involved clauses in the EPC ITB and contracts by the phrase-matcher technique. A machine learning model was built in the TFA module for contractual risk extraction by using the named-entity recognition (NER) method. The risk-involved clauses collected for model development were converted into a database in JavaScript Object Notation (JSON) format, and the final results were saved in pickle format through the digital modules. In addition, optimization and reliability validation of these modules were performed through Proof of Concept (PoC) as a case study, and the modules were further developed to a cloud-service platform for application. The pilot test results showed that risk clause extraction accuracy rates with the CRC module and the TFA module were about 92% and 88%, respectively, whereas the risk clause extraction accuracy rates manually by the engineers were about 70% and 86%, respectively. The time required for ITB analysis was significantly shorter with the digital modules than by the engineers. [ABSTRACT FROM AUTHOR]