Big data analytics : 10th international conference, BDA 2022, Hyderabad, India, December 19-22, 2022 : proceedings

This book constitutes the proceedings of the 10th International Conference on Big Data Analytics, BDA 2022, which took place in Hyderabad, India, in December 2022. The 7 full papers and 7 short papers presented in this volume were carefully reviewed and selected from 36 submissions. The book also contains 4 keynote talks in full-paper length. The papers are organized in the following topical sections: Big Data Analytics: Vision and Perspectives; Data Science: Architectures; Data Science: Applications; Graph Analytics; Pattern Mining; Predictive Analytics in Agriculture. Preface Organization Contents Big Data Analytics: Vision and Perspectives Data Challenges and Societal Impacts – The Case in Favor of the Blueprint for an AI Bill of Rights (Keynote Remarks) 1 Introduction 2 Benefits of AI 3 AI Bias 3.1 Statistical Bias 3.2 Human and Systemic Bias 3.3 Beyond Bias - The Issue of Consent 4 A Framework to Build Trustworthiness 4.1 Technical Characteristics 4.2 Socio-Technical Characteristics 5 Conclusion 5.1 Need for Research 5.2 Need for Legislation References Big Data in Cognitive Neuroscience: Opportunities and Challenges 1 Introduction 1.1 Functional Segregation and Functional Integration 2 Inferential Approaches in Cognitive Neuroscience 3 Current Practices in Cognitive Neuroscience 4 Opportunities 5 Challenges 6 Conclusion References Data Science: Architectures A Novel Feature Selection Based Text Classification Using Multi-layer ELM 1 Introduction 1.1 Research Motivation 1.2 Research Contribution 2 Prelims 2.1 Multi-layer ELM 3 Methodology 4 Analysis of Experimental Results 4.1 Experimental Setup 4.2 Discussion 4.3 Comparisons of ELM and ML-ELM Feature Space 5 Conclusion References ARCORE: A Requirements Dataset for Service Identification 1 Introduction 2 Related Work 2.1 Requirements Datasets: 2.2 Service Selection 2.3 Requirements Classifications 2.4 Techniques Used for Automatic Requirements Classification 3 ARCORE Dataset 3.1 Service Cues Creation 3.2 Validation and Annotation Guide Creation 3.3 Requirement Corpus Creation 3.4 ARCORE Dataset Creation 3.5 Sample Response Explanation 4 Conclusion and Future Work References Learning Enhancement Using Question-Answer Generation for e-Book Using Contrastive Fine-Tuned T5 1 Introduction 2 Background 3 Methodology 3.1 T5 - Abstractive Summarizer 3.2 Edu Question-Answer Generation (eQAG) 4 Experimental Results and Discussion 4.1 Dataset 4.2 Model Evaluation 4.3 Comparison with Baselines 4.4 Human Evaluation for Relevancy Testing 5 Conclusion and Future Scope A BERT Score for Semantic Match B Few More Examples of Generated QAs for Text Document References Data Science: Applications A Machine and Deep Learning Framework to Retain Customers Based on Their Lifetime Value 1 Introduction 2 Related Work 3 Methodology 4 Design Specification 4.1 Customer Segmentation Models 4.2 Customer Lifetime Value Prediction Models 5 Implementation 6 Evaluation 6.1 Evaluation of Segmentation 6.2 Evaluation of Customer Lifetime Value Models 6.3 Discussion 7 Conclusion and Future Work References A Deep Learning Based Approach to Automate Clinical Coding of Electronic Health Records 1 Introduction 2 Related Work 3 Presented Automated Clinical Coding Models 3.1 ICD-9 Codes 3.2 Presented Models 3.3 Baseline Word2vec and Cosine Similarity Hybrid Model 3.4 Transformer Encoder Model Results 3.5 BERT Model (BlueBERT) 4 Experimental Analysis and Results 4.1 Used MIMIC-III Dataset 4.2 Evaluation Metrics 4.3 Implementation Details 4.4 Results and Analysis 5 Conclusion References Determining the Severity of Dementia Using Ensemble Learning 1 Introduction 2 Literature Review 3 Proposed Multi-phase Detection of Dementia 3.1 Phase 1 - Dementia Detection Using ADL Data 3.2 Phase 2 - Dementia Severity Prediction Using MRI Scans 3.3 Application of Random Forest Classifier in Phase 1 and 2 of Dementia Detection 4 Experimental Study 4.1 Analysis on Phase 1 Using ADL Data 4.2 Analysis of Phase 2 Using MRI Data 5 Conclusion References A Distributed Ensemble Machine Learning Technique for Emotion Classification from Vocal Cues 1 Introduction 2 Related Works 3 Proposed Framework 3.1 Dataset 3.2 Preprocessing 3.3 Feature Extraction and Reduction 3.4 Distributed Machine Learning Algorithms 4 Experimental Setup and Analysis 4.1 Results 5 Conclusion References Graph Analytics Drugomics: Knowledge Graph & AI to Construct Physicians’ Brain Digital Twin to Prevent Drug Side-Effects and Patient Harm 1 Introduction 2 Drug-Drug Interaction (DDI) Knowledge Sources 3 Drug-Disease Interaction (DDSI) Knowledge Sources 4 Drugomics Knowledge Graph 5 Drugomics Use Case with Clinical Decision Support 5.1 Chief Complaints 5.2 Provisional Diagnosis 5.3 Prescription 5.4 Primary Diagnosis 5.5 Drugomics Interactions 6 Conclusion References Extremely Randomized Tree Based Sentiment Polarity Classification on Online Product Reviews 1 Introduction 2 Related Works 3 Methodology 3.1 Data Set 3.2 Text Pre-processing 3.3 Feature Extraction 3.4 Unigram Model 4 Classification 4.1 Ensemble Methods 4.2 Base Classifiers 4.3 Performance Evaluation Parameters 5 Result and Discussion 6 Conclusion References Community Detection in Large Directed Graphs 1 Introduction 2 Related Work 3 Our Approach 3.1 PageRank 3.2 Overall Algorithm Outline 3.3 Time Complexity 4 Experiments and Results 4.1 Nine-Level Communities 4.2 Community Coefficient and Community Size 4.3 Effect of PageRank Threshold k 4.4 Scalability Analysis 5 Conclusions References Pattern Mining FastTIRP: Efficient Discovery of Time-Interval Related Patterns 1 Introduction 2 Problem Definition 3 The FastTIRP Algorithm 3.1 The Search Process 3.2 The Pair Support Pruning Technique 4 Experimental Evaluation 4.1 Influence of minsup on Runtime, Number of Joins and Patterns 4.2 Influence of minsup on the Overall Memory Usage 5 Conclusion References Discovering Top-k Periodic-Frequent Patterns in Very Large Temporal Databases 1 Introduction 2 Related Work 3 Proposed Model: top-k Periodic-Frequent Patterns 4 Our Algorithm 4.1 Basic Idea: Dynamic Maximum Periodicity 4.2 k-PFPMiner 5 Experimental Results 5.1 Experimental Setup 5.2 Evaluation of Algorithm by Varying only k 5.3 Scalability Test 6 Conclusions and Future Work References Hui2Vec: Learning Transaction Embedding Through High Utility Itemsets 1 Introduction 2 Related Work 3 Framework 3.1 Problem Definition 3.2 Learning Transaction Embeddings Based on Items 3.3 Learning Transaction Embedding Based on High Utility Itemsets 3.4 Hui2Vec Methods to Learn Transaction Embeddings 4 Experiments 4.1 Datasets 4.2 Implemented Models 4.3 Evaluation Metrics 4.4 Parameter Settings 4.5 Results and Discussion 5 Conclusion References Predictive Analytics in Agriculture A Data-Driven, Farmer-Oriented Agricultural Crop Recommendation Engine (ACRE) 1 Introduction 1.1 Motivation for ACRE 1.2 Contributions and Outline 2 Review of Relevant Work 2.1 Relevant Work in Crop Recommendation Systems 2.2 Relevant Work in Crop Yield Prediction 2.3 Positioning of Our Work 3 Sharpe Ratio 4 Data Collection and Curation 4.1 Yield Data 4.2 Weather Data 4.3 Soil Data 5 Building Blocks of ACRE 5.1 Input Parameters 5.2 Utility Calculator 6 Experiments and Results 6.1 Crop Yield Prediction 6.2 Results on Profit Utilities 6.3 Recommendation of Individual Crops 6.4 Sharpe Ratio Based Crop Portfolio Recommendation 6.5 Socio-Cultural Factors in Crop Recommendation 7 Summary and Future Work References Analyze the Impact of Weather Parameters for Crop Yield Prediction Using Deep Learning 1 Introduction 2 Related Work 3 Dataset and Methods 3.1 Study Area 3.2 MODIS Image Datasets 3.3 Weather Data 3.4 Proposed Method 4 Result and Discussion 4.1 Model’s Performance 4.2 Comparison with Other Models 5 Conclusion References Analysis of Weather Condition Based Reuse Among Agromet Advisory: A Validation Study 1 Introduction 2 Materials and Methods 2.1 About Agromet Advisory Service 2.2 A CWC-based Reuse Framework 2.3 Methodology 2.4 Experimental Setup 3 Results and Discussion 3.1 Cluster Analysis of CWCs 3.2 Cluster Analysis of Advisory Data 3.3 Discussion 4 Conclusion References Author Index