Search

Due to the increasing use of information technologies by biomedical experts, researchers, public health agencies, and healthcare professionals, a large number of scientific literatures, clinical notes, and other structured and unstructured text resources are rapidly increasing and being stored in various data sources like PubMed. These massive text resources can be leveraged to extract valuable knowledge and insights using machine learning techniques. Recent advancement in neural network-based classification models has gained popularity which takes numeric vectors (aka word representation) of training data as the input to train classification models. Better the input vectors, more accurate would be the classification. Word representations are learned as the distribution of words in an embedding space, wherein each word has its vector and the semantically similar words based on the contexts appear nearby each other. However, such distributional word representations are incapable of encapsulating relational semantics between distant words. In the biomedical domain, relation mining is a well-studied problem which aims to extract relational words, which associates distant entities generally representing the subject and object of a sentence. Our goal is to capture the relational semantics information between distant words from a large corpus to learn enhanced word representation and employ the learned word representation for various natural language processing tasks such as text classification. In this article, we have proposed an application of biomedical relation triplets to learn word representation through incorporating relational semantic information within the distributional representation of words. In other words, the proposed approach aims to capture both distributional and relational contexts of the words to learn their numeric vectors from text corpus. We have also proposed an application of the learned word representations for text classification. The proposed approach is evaluated over multiple benchmark datasets, and the efficacy of the learned word representations is tested in terms of word similarity and concept categorization tasks. Our proposed approach provides better performance in comparison to the state-of-the-art GloVe model. Furthermore, we have applied the learned word representations to classify biomedical texts using four neural network-based classification models, and the classification accuracy further confirms the effectiveness of the learned word representations by our proposed approach.

In recent years, the application of various recommendation algorithms on over-the-top (OTT) platforms such as Amazon Prime and Netflix has been explored, but the existing recommendation systems are less effective because either they fail to take an advantage of exploiting the inherent user relationship or they are not capable of precisely defining the user relationship. On such platforms, users generally express their preferences for movies and TV shows and also give ratings to them. For a recommendation system to be effective, it is important to establish an accurate and precise relationship between the users. Hence, there is a scope of research for effective recommendation systems that can define a relationship between users and then use the relationship to enhance the user experiences. In this research article, we have presented a hybrid recommendation system that determines the degree of friendship among the viewers based on mutual liking and recommendations on OTT platforms. The proposed enhanced model is an effective recommendation model for determining the degree of friendship among viewers with improved user experience.

There is a considerable rise in cardiovascular diseases in the world. It is pertinently essential to make cardiovascular prediction accurate to the maximum. A forecast based on machine learning techniques can be beneficial in detecting cardiovascular disease (CVD) with maximum precision and accuracy. The disease's effective prediction helps in early diagnosis, which cuts down the mortality rate. A health history and the causes of heart disease require the efficient detection and prediction of CVD. Data analytics is beneficial for making predictions based on a massive amount of data, and it aids health clinics in disease prognosis. Regularly, a large volume of patient-related data is maintained. The information gathered can be used to forecast the emergence of upcoming diseases. Our study presents a detailed comparative study of Cardiovascular Disease by comparing the various machine learning techniques mainly comprising of classification and predictive algorithms. The study shows an in-depth analysis of around forty-one papers related to cardiovascular disease by using machine learning techniques. This study evaluates the selected publications rigorously and identifies gaps in the available literature, making it useful for researchers to develop and apply in clinical fields, primarily on datasets related to heart disease. The current study will aid medical practitioners in predicting heart threats ahead of time, allowing them to take preventative measures.

Murkiness in pictures is because of common ecological wonders, which makes the picture in a white shade commotion. Fog evacuation is one of the most significant research subjects nowadays to the due prevalence of utilizations progressively reconnaissance from rambles or any region under security. Both indoor and outside pictures are significant for testing cloudiness and its expulsion. Many picture-handling methods are made by analysts to evacuate cloudiness in a solitary picture. Murkiness force can be determined by a parameter known as perceptual fog density measure (PFD). It is critical to investigate this parameter for all the strategies in order to get a thought of progress. In this postulation, another methodology is made by applying an all-inclusive guided sifting procedure with a profound neural system. This proposed calculation is executed on MATLAB programming, and results are gotten by figuring the PFD in the current and proposed method. The four methods are compared with one another. The methods are global image filtering (GIF), weighted global image filtering (WGIF), globally guided image filtering (GGIF), and a proposed strategy, for example, comprehensively guided sifting with deep neural network (DNN). In GIF, the fine structure of the picture is commonly not safeguarded and an unreasonable picture is acquired. In WGIF, the PFD got is most noteworthy. In GGIF, PFD is lower and structure is not safeguarded, yet in the proposed calculation, the PDF is least with fine structure, and shading power of the image is of the best quality.

The advancement in artificial intelligence has led to the modernization of many fields including games, sports and shopping experiences but the judicial system has much more potential in overcoming some of the basic needs which is time and money. The modern techniques which can be used to solve some of the most complex problems like AlphaGo and other protein predicting algorithms to fight against cancer are not any more a fantasy but a reality. And with all these latest developments, the judicial system has more opportunity to grab and make a better society. But using the CNN and clustering, we can make a simpler approach to solve the large pending cases in the lower courts in Indian judicial system. In this paper, there has been an approach to make a way to use the techniques and find a way to solve the problem of pending cases in the lower court.

The COVID-19 pandemic has emerged as one of the most crucial health emergencies in the last decade where almost all entities of a nation?s ecosystem like inhabitants, businesses, governments, economies, and environment are impacted. The large volumes of epidemiological, clinical, personal, and environmental data generated during any pandemic can provide useful insights about the underlying causes, symptoms, relations, and correlations, which if analyzed can assist in mitigating the impact to a great extent. The cheap and easy connectivity and communication provided by the social media platforms (SMP) have established them as one of the most preferred mediums of communications among the masses. The data generated by these platforms can be analyzed in context of the ongoing COVID-19 crisis to provide critical information and insights related to the ground level realities like spread and severity of infection, the state of implementation of control measures, the mental state of individuals, and their needs. The tweets and comments of the users can provide information about the current situation and intensity of the problems in the affected regions. With the help of techniques like sentiment analysis and web mining, we can identify the emergent requirements and needs (like food, shelter, medicine, medical emergencies, security, etc.) of the population in the COVID-19-affected areas. With this chapter we aim to identify several use cases where the big medical data from the patients, epidemiological data, social media data, and environment-related data can be used to identify patterns, causes, and other growing factors of the COVID-19 pandemic with a goal to mitigate the damages and contain further spread of the disease. The chapter also discusses the impact of a preferred mitigation measure of nationwide lockdown on the number of new novel coronavirus-positive patients as well as the impact on the environment by analyzing the available data. Since the tourism industry is now of the worst hit businesses, we also discussed the impact of COVID-19 on tourism industry. Furthermore, we identify the challenges associated with handling the massive amount of data generated during such pandemics. Finally, the future avenues of using big data for effectively devising predictive mechanisms and techniques to contain such pandemics in the initial stages are discussed. The chapter also discusses the importance of edge/fog technologies and IoT to identify possible use cases and where immediate point of contact actions is needed to mitigate the situations. Since edge computing facilitates calculations near the origin of data, it is imperative to understand the potential use cases in times of COVID-19-like pandemics. 5311-8814-F0ED | Sara Maria da Cruz Maia de Oliveira Paiva N/A

Data mining is the technique of extracting relevant information from large heterogeneous data sources and present in a meaningful and systematic way for users. Nowadays, data mining has become a trend and all computer-related industries are engaged with data mining. Currently, many data mining algorithms focused on computing the frequent pattern from certain and uncertain databases. In this paper, a new algorithm named FuzzyUeclat is being proposed which uses Fuzzy operators to extract frequent items from the vertical probabilistic dataset. Previously, the frequent items were computed on the basis of expected minimum support derived from the existential probability associated with items. Therefore, as the size of the database increases, the value of expected support is decreased at each level of iterations and subsequently, it approaches zero. To avoid the loss of frequent itemset, the expected minimum support is being replaced with Fuzzy min and Fuzzy max operator to compute the support for an itemset. The experimental result shows that the algorithm FuzzyUeclat increases the frequent patterns and improves the performance for the large uncertain databases.

In this paper a new methodology is described to efficiently handle and managed traffic in a highly populated and congested area. The traffic management system’s framework make use of an essential technology required which is IoT .It also has other important parts such as an autonomous activity controller raspberry-pi, pi-camera, RFID ,IR sensors .To further help to navigate the traffic the proposed framework uses decisive algorithm and round-robin algorithm to find the Optimum path through traffic. Raspberry pi is used to manage all components collectively and individually. Total traffic in a region is determined by the help of IR sensors which uses RFID to distinguish high priority vehicles such as ambulance and VIPS from day to day traffic. There is an additional benefit of using RFID it can be used to locate the robbed or snatched vehicles. The proposed framework has two major parts i.e autonomous and manual. The algorithm used to navigate traffic play a vital part in making the proposed framework work efficiently.

Data mining is a technique through which we can find interesting data and sequence from the available wide range of data source. Incremental high-average utility pattern mining (IHAUPM) algorithm is represented to manage the incremental database with transaction insertion. IHAUPM algorithm basically follows the comparison to the original database and newly inserted database itemset if itemset has High Average Utility Upper Bound Itemset (HAUUBI) in the initial database as well as new transaction database then the item always frequent. Second situation itemset has non-High Average Utility Upper Bound Itemset (non-HAUUBI) in the initial database as well as new transaction database then the item always not frequent. Otherwise, the itemset is recurring or not is identified by the given information. This new algorithm is represented in this paper to generate expected high utility frequent item set to form a new transaction database; this algorithm is much faster than the existing algorithm.

systems have become extremely popular in the last decade and are applied to various e-commerce websites. Recommender systems help users find useful, interesting items or content from a considerably large amount of information. The major challenge in recommender system is Cold-start Problem (1), this appears when a user or item is new, the system may fail because not enough information is available for this user or item. There are so many solutions are proposed to solve cold-start problem including matrix factorization, graph based (2) (3) etc. One more serious form of cold-start found in some real-time e-commerce applications that is named as CoCoS (Continuous Cold Start) problem(4), this is a recurring version of cold-start even for known users or items, since many users visit the website rarely, change their interests in time, or exhibit different personas. Hence both of the widely used approaches CBF (Content Based Filtering)(5) and CF(Collaborative Filtering)(6) will suffer from this problem. The most basic assumption -"similar users will like similar items" (7) of Recommender System fails in some cases where user interest is changing over time and hence CoCoS problem arises. CoCoS problem is domain specific not all e-commerce website falls under CoCoS, but there are some websites that suffer from CoCos due to the changes in user's interest over time, the best example is travel and tourism domain based websites. Users booked a ticket or package on their current needs and interest by using different websites and we won't be able to make a useful set of suggested items because we don't have sufficient information on user's recent activities or requirements to generate a set of recommended items. In this paper, we are proposing a solution to get the recent preferences and interest of user on the basis of Social Network Activities (SNA). Here we are targeting travel and tourism domain to state the CoCoS problem and to find out the solution of CoCoS by using SNA (Social Network Activities) . Here we are considering Facebook (8) Check-ins Count (FCC) for obtaining the information about user's recently visited places and to draw a conclusion for user's preferences for next visit.

Simplification is an integral part of Machine translation, however, it still remains the most complex part of the process. A sentence in Hindi can be written in multiple ways. They can be complex sentences or simple ones. These sentences then need to be translated into English language. For this, the complex sentences need to be converted into simple sentences before being translated. This paper concerns Sentence Simplification of Complex Hindi Sentences for the Hindi shallow parser that was developed by IIIT. Complexity of a sentence can occur due to presence of clauses, multiple verbs and usage of conjunctions. So splitting the sentence works if the above mentioned properties of the sentence can be eliminated. To automate the process, a simplification algorithm has been formed. The paper shall talk about errors and patterns that have been analyzed, so as to improve the accuracy of the simplified sentence, and preserving the sense of the original sentence. General Terms Computational Linguistics, Natural Language Processing, Sentence Simplification, Shallow Parser, Translation.

In this paper, we have proposed a novel concept to optimise ordered weighted aggregation (OWA) based fuzzy time series predictor (FTSP) using genetic algorithm (GA). Firstly, accurateness of FTSP is enhanced by applying effective method of aggregation on past observations using OWA weights. These weights are determined on the basis of importance of fuzzy set in the system by employing regularly increasing monotonic (RIM) quantifiers. Subsequently, GA is used to optimise membership functions of FTSP by generating its wide range of parameters in the region of time series. Lastly, this model is capable of controlling its performance by varying GA parameters. To assess proposed method, we used dataset of enrolments and outpatient visits, as used by almost all previous research in this domain. Evaluation results indicate coalescing OWA and GA for FTSP significantly reduced mean square error (MSE) and average forecasting error rate (AFER).