Framework for Evaluating Feature Selection in Opinion Mining

Text mining is a special case of data mining which explore unstructured or semi-structured text documents, to establish valuable patterns and rules that indicate trends and significant features about specific topics. Text mining has been in pattern recognition, predictive studies, sentiment analysis and statistical theories in many areas of research, medicine, financial analysis, social life analysis, and business intelligence. Text mining uses concept of natural language processing and machine learning. Machine learning algorithms have been used and reported to give great results, but their performance of machine learning algorithms is affected by factors such as dataset domain, number of classes, length of the corpus, and feature selection techniques used. Redundant attribute affects the performance of the classification algorithm, but this can be reduced by using different feature selection techniques and dimensionality reduction techniques. Feature selection is a data preprocessing step that chooses a subset of input variable while eliminating features with little or no predictive information. Feature selection techniques are Information gain, Term Frequency, Term Frequency-Inverse document frequency, Mutual Information, and Chi-Square, which can use a filters, wrappers, or embedded approaches. To get the most value from machine learning, pairing the best algorithms with the right tools and processes is necessary. Little research has been done on the effect of feature selection techniques on classification accuracy for pairing of these algorithms with the best feature selection techniques for optimal results. In this research, a text classification experiment was conducted using incident management dataset, where incidents were classified into their resolver groups. Support vector machine (SVM), K-Nearest Neighbors (KNN), Naïve Bayes (NB) and Decision tree (DT) machine learning algorithms were examined. Filtering approach was used on the feature selection techniques, with different ranking indices applied for optimal feature set and classification accuracy results analyzed. The classification accuracy results obtained using TF were, 88% for SVM, 70% for NB, 79% for Decision tree, and KNN had 55%, while Boolean registered 90%, 83%, 82% and 75%, for SVM, NB, DT, and KNN respectively. TF-IDF, had 91%, 83%, 76%, and 56% for SVM, NB, DT, and KNN respectively. The results showed that algorithm performance is affected by feature selection technique applied. SVM performed best, followed by DT, KNN and finally NB. In conclusion, presence of noisy data leads to poor learning performance and increases the computational time. The classifiers performed differently depending on the feature selection technique applied. For optimal results, the classifier that performed best together with the feature selection technique with the best feature subset should be applied for all types of data for accurate classification performance. Keywords: Text Classification, Supervised Machine Learning, Feature Selection DOI : 10.7176/JIEA/9-3-06 Publication date :May 31 st 2019

Reviews of women's tops in the market are valuable information if processed properly. Merchants can conduct product review analysis to obtain information that can be used to evaluate products and services. Product review analysis activities are not enough just to see the number of stars, it is necessary to see the entire content of the review comments to be able to know the intent of the review. Sentiment analysis system is a system used to automatically analyze online product reviews to obtain information including sentiment information that is part of online reviews. The data is classified using Naive Bayes. The data collected were 1,000 product reviews of women's tops as samples. The purpose of this study is to determine the sentiment analysis of female top product reviews using the Naive Bayes algorithm. The stages of this research include data collection, labeling, pre-processing, sentiment classification, and evaluation. In the pre-processing stage there are 6 stages, namely Cleaning emoticons & symbols, Case folding, Word Normalizer, Tokenize, Stopword Removal and Stemming. TF-IDF (Term Frequency - Inverse Document Frequency) method is used for word weighting. The data will be classified into 3 (three) classes, namely negative, positive and neutral. The data will then be evaluated using accuracy parameter testing. The test results show an accuracy value of 89%, this result shows that the product reviews of women's tops are more positive.

Citation (2023), "Index", Tyagi, P., Grima, S., Sood, K., Balamurugan, B., Özen, E. and Eleftherios, T. (Ed.) Smart Analytics, Artificial Intelligence and Sustainable Performance Management in a Global Digitalised Economy (Contemporary Studies in Economic and Financial Analysis, Vol. 110B), Emerald Publishing Limited, Bingley, pp. 265-276. https://doi.org/10.1108/S1569-37592023000110B018 Publisher: Emerald Publishing Limited Copyright © 2023 Pallavi Tyagi, Simon Grima, Kiran Sood, B. Balamurugan, Ercan Özen, and Thalassinos Eleftherios INDEX Accounting, 146 Actual economic activity, 175 Administrative security controls in SME, current state of, 35–37 Africa, effect of COVID-19 debt accumulation, 165 Agriculture, 133 Alternative finance models, 102 Amazon (technology company), 98 Annual appraisals, 252 Anti-black Money Act (2015), 126 Anti-money laundering (AML), 60, 172 Apple (technology company), 98 Artificial intelligence (AI), 23, 85–86, 172, 217, 244, 252–253 ability of AI and SA to improve PM systems, 257 application of AI and SA in business processes, 255 automating PM Systems, 259–261 in business processes, 255–256 classification models to predict employee performance, 257–259 classification of AI, based on capabilities, 253–254 classification of AI, based on functionality, 254 competition, 219 demographics of customers, 219 drivers of digitisation and AI in retail, 218 in HR management, 259 limitations of AI and SA in PM systems, 261–262 for mapping cybersecurity controls for SME, 46–48 in PM systems, 257 retail customers, 218–219 retail digitisation using, 219–220 super, 254 technology and, 218 types, 253 Aspirational Districts Programme, The, 88 Asset management segment, 99 Augmented reality, 23 Authors’ network analysis, citation as per, 68–69 Autonomous robotics, 23 Balance sheet lending (BSL), 100 Bank 4.00, 102 Bank of Baroda, 7, 11–12 Bank of Hindustan, 2 Banking, 2, 98, 206 co-operative banks, 2 commercial banks, 2 industry, 205 payments banks, 2 small finance banks, 2 Banking, financial services, and insurance (BFSI), 29 Banking Regulation Act (1949), 2 Banks, 205 Bibexcel, 135 Bibliometrics, 57, 60 analysis, 141 of entrepreneurial universities, 57 R Package, 135 research, 141 Bibliometrix, 178 Biblioshiny, 178 Big data, 23, 85, 255 analytics, 84 BigTechs, 101 Billion Prices Project, 87 Black box model, 85 Black money, 58–59 Blending data sciences in economic and social issues, 86–87 Blockchain, 132 content analysis, 137–141 data analysis, 135–137 directions for future research, 141–142 implications, 141 methodology, 133–134 technology, 132 Bombay Stock Exchange (BSE), 75–76 Bootstrapping methods, 227 Bring your own device policy (BYOD policy), 40 Business domain-wise mission critical asset security (Business DMCAS), 43–44 Business to-Person lending models (B2P lending models), 101 Business-to-Business lending models (B2B lending models), 101 Businesses, 230 AI in, 255–256 changing priorities of CIA triad based on business domain, 30–31 SA in, 256–257 Buy Now Pay Later (BNPL), 101 C4.5 decision tree, 258 Canara Bank, 7, 9–10, 18 Capability maturity model (CMM), 26 Cashkumar (P2P lending platform), 107 version 2. 0, 108 Cashkumar Peer-to-Peer Lending Platform (Cashkumar P2P Lending Platform), 107–108 digital lending, 100–107 emotional analysis, 116–117 FinTech, 98–100, 108–109 literature review, 108 negative reviews, 119–120 P2P lending, 109–111 positive reviews, 118–119 research methodology, 112 sentiment analysis perspective, 111–112 theoretical background, 107 Castle approach, 28 Casual conversations, 250 Central Bank, 193 Central bank digital currency (CBDC), 190 Central Bank of Nigeria, 190 CIA triad assess priority of CIA triad for DMCA, 43 changing priorities of CIA triad based on business domain, 30–31 CIA trinity, 27 Citation network analysis, 67–68 citation as per authors’ network analysis, 68–69 Cloud computing, 23, 86 Co-authorship analysis, 60 Co-authorship network analysis as authors on money laundering, 62–66 as countries, 64–66 Co-occurrences analysis, 60 of keywords analysis on money laundering, 66–69 network analysis on money laundering, 66 Co-operative banks, 2, 4 Commercial banking, 2 Commercial banks, 2 foreign banks, 3 private sector banks, 3 public sector banks, 3 RRBs, 3 Communications service providers, 101 Companies Act, 105 Competitive pressure, 229 Conceptual model constructs and items measured, 221–223 development of, 221 Conceptual research model, 221 Confidence interval (CI), 227 Consumers, 216 Consumption of goods and services, 22 Content analysis, 137, 179 digitisation of agriculture, 140 supply chain management, 140 sustainable economic development, 140–141 Cooperative savings, 59 Cooperative Societies Act (1912), 4 Core cybersecurity concepts, 26–31 changing priorities of CIA triad based on business domain, 30–31 CIA Triad, 27 Defence in Depth, 28–29 DMCA, 29–30 Coronavirus, 162 Corporation threat agents, 24 Corruption, 56, 124 Cost-benefit analysis, 84 COVID 19, 5, 218 Africa, 165 effect of COVID-19 debt accumulation, 164 debt crisis, 164 debt statistics, 162–164 Europe, 164–165 global debt crisis, 161 heavily indebted countries, 167–168 and Indian Banking, 5 literature review, 161–162 methodology, 161 multilateral organisations, 166–167 pandemic, 160, 234–235 rich countries, 165–166 solution to overcome, 165 Cross industry standard process for data mining (CRISP-DM), 257 Crowdfunding, 99–100 Cryptocurrencies, 98 Customer relationship management (CRM), 253 Customers, 229 demographics of, 219 screening process, 58 Cutting-edge technology, 58 Cyber assets, 27 Cyber risk, 23 Cyberattacks faced by SMEs, experience of, 39–40 Cybercrime, 56 Cybercriminals, 37, 40 Cybersecurity, 23 calculate SME’s minimum overall cybersecurity controls implementation level, 44–45 implement prioritised cybersecurity controls for Business DMCAS, 43–44 responsible AI for mapping cybersecurity controls for SME, 46–48 Cybersecurity framework (CSF), 25 Cyberspace, 23 Data analysis, 224 measurement model, 224–226 structural model, 226–229 test, 224 tool, 88 Data collection, 223–224 Data extraction, 134 Data lags, 91–92 Data localisation problem, 92 Data mining, 85 techniques, 90 Data privacy and security, 91 Data science, 86 techniques, 85 Data-driven policy, India’s stance on, 87–89 Debt crisis, 161, 164 Debt restructuring, 162 Debt Service Suspension Initiative (DSSI), 166 Debt statistics, 162 external debt stocks, 162–163 massive spending by richer countries, 163–164 sovereign default, 164 Decision Tree, 257 Decision-making management, 173 Defence in depth (DiD), 28 Delay payments, 204 Democratisation of opinions, 236 Descriptive analysis, 60 Desktops, 40 Digital circuits, 22 Digital government, 87 Digital India campaigns, 89 Digital lending, 100 alternative finance models, 102 ecosystem, 100–102 functional business models in India, 105–107 funds transfer mechanism as per RBI guidelines, 105 global scenario, 101 governing enactments, 104–105 India, 102 landscape, 102–104 market share, 102 Digital sentiments, 236 Digital technologies, 109 Digital transformation, 216 in agriculture, 141 Digitalisation process of banking industry, 6 Digitisation intelligence competition, 219 demographics of customers, 219 drivers of digitisation and AI in retail, 218 retail customers, 218–219 technology and AI, 218 Digitisation of agriculture, 133, 140 Digitisation transforms retailing exchanges, 220 Directorate of Enforcement (ED), 126 Dirty money, 124 Document-term matrix (DTM), 178 Domain-wise Least Cybersecurity Implementation Framework (DLCI Framework), 41 assess priority of CIA triad for DMCA, 43 calculate SME’s DLCI Level, 45–46 calculate SME’s DMCAS Level, 44 calculate SME’s minimum overall cybersecurity controls implementation level, 44–45 identify DMCA, 41–43 implement prioritised cybersecurity controls for business DMCAS, 43–44 responsible AI for mapping cybersecurity controls for SME, 46–48 WCCI for SME, 44 Domain-wise Mission Critical Asset (DMCA), 29–30 Domestic lending by banks, 205 Drug trafficking, 56 e-CNY, 192 E-commerce platforms, 101 sites, 216 Earnings management, 146 conceptual framework, 147 income-targeting approach, 150–151 literature review, 148–150 priority theory of sustainable finance, 147–148 surplus income model, 152–155 surplus income model for sustainability, 151–152 Econometric models, 84 Economic and financial performance in topic modelling framework, relationship between money laundering and, 184–185 Economic growth (EG), 125, 128, 161, 173, 201 Electric vehicles infrastructure, investment target for, 202 Electronic word of mouth (e-WOM), 235 Embed quality assurance, 47 Emotion Lexicon (EmoLEX), 111 Emotional analysis, 116–117 Emotions, 116 Employees, 245, 258 in India, 175 narrow assessment of, 252 performance, 245 threat agents, 24 eNaira, 190, 193–195 CBDC, 190 money, 190 performs, 190 speed wallet, 196 Environmental, social and governance (ESG), 76, 79–81 ESG-based socially responsible investments, 75 Indian investors looking for, 75–76 Environmental accounting, 149 Environmental reporting, 146 Europe, effect of COVID-19 debt accumulation, 164–165 European Central Bank (ECB), 162 External debt stocks, 162–163 External incentives, 154 External stakeholders, 248 Facebook (technology company), 98 Federal Bank, 7, 13–14, 18 Finance models, 102 Financial Action Task Force (FATF), 58, 125 Financial crimes action plan for resolving financial crimes using SRI–ESG model, 81 channel for, 77–78 effects of, 79 terrorism financing devastating part of, 78–79 impact of unethical investments on, 79 Financial crisis (2008), 98 Financial industry, 205 Financial institutions, 205 in achieving sustainable economic goals, 206–207 through priority sector lending, 208–210 Financial markets, 101, 104 Financial performance Bank of Baroda, 11–12 Canara Bank, 9–10 co-operative banks, 4 COVID 19 and Indian Banking, 5 COVID 19, 5 data analysis, 8 data collection, 7 Federal Bank, 13–14 findings, 17–18 foreign banks, 3 HDFC Bank, 14–15 hypothesis testing for net profit, 16–17 hypothesis testing for NPA, 17 ICICI Bank, 12–13 Indian Bank, 10–11 limitation of study, 8 literature review, 5–7 non-performing assets, 4 non-scheduled banks, 4 payments banks, 4 private sector banks, 3 public sector banks, 3 research methodology, 7 research methodology, 8 research objectives, 7 RRB, 3 sample frame, 7 schedule banks, 4 small finance banks, 3–4 in topic modelling framework, relationship between money laundering and economic and, 184–185 Union Bank, 10 Yes Bank, 15–16 Financial services, 98 Financial study, 58 Financial Technology (FinTech), 98 branches of FinTech Industry, 99–100 classified into the following types, 98–99 credit, 102 landscape, 99 lenders, 101 start-ups, 109 Financing segment, 99 Firms, 147 Five-year planning process, 87 Forecasting methods, 84 Foreign Banks, 3 Formal process in PM systems, 249–250 Fugitive Economic Offenders Act (2018), 126 Functional business models in India, 105–107 Funds transfer mechanism as per RBI guidelines, 105 Global economy, 160 Global reporting initiative (GRI), 149 Globalisation of economies, 124 Gold medal, 238 Goods and Service Tax, 126 Google (Private companies), 91, 98 reviews, 114 Governance in India, 87 reporting, 146 Grameen Foundation, The, 87 Green accounting, 149 Green banking assessing funding needed by international agency, 205 investment target for electric vehicles infrastructure, 202 investment target for renewable energy generation, 201–202 investment target to set up smart cities, 203–204 methodology, 206 results, 206 role of financial institutions in achieving sustainable economic goals, 206–207 role of financial institutions through priority sector lending, 208–210 Green development, 210 Green energy, 210 Green growth, 206 Gross domestic product (GDP), 160, 174, 205 HDFC Bank, 7, 14–15, 18 Human bias, 252 Human resources (HR), 172, 255 AI in HR management, 259 Human-based PM systems, 252 Hume’s Aesthetic Theory, 108 Hypothesis testing for net profit, 16–17 for NPA, 17 ICICI Bank, 7, 12–13, 18 ID3 decision tree, 258 Illicit drug trafficking, 59 IMF Fact Sheet, 125 Income smoothing, 151 Income-targeting approach, 150–151 Incremental approach, 84 India current market trend of, 74–75 digital lending, 102 functional business models in, 105–107 measurements to control money laundering terror funding in, 125–126 retail scenario in, 217–218 SRI in, 75 stance on data-driven policy, 87–89 Indian Bank, 7, 10–11 Indian banking industry, 4 sector, 2 Indian Economy future implication of study, 128 impacts of money laundering on, 126–127 impacts of terror financing on, 127–128 measurements to control money laundering terror funding in India, 125–126 nexus of money laundering and terrorism financing, 125 Indian government, 90 Indian investors looking for ESG Investments, 75–76 Industry 1.0, 22 Industry 2.0, 22 Industry 4.0 digitisation, 23 Inferential research approach, 223 Informal economy, 174 Informal process in PM systems, 250 Information and communications technologies (ICTs), 23, 88, 217 assets, 27 Information security, 27 Information security management system (ISMS), 24 Information technology, 124 Institute for Energy Economics and Financial Analysis (IEEFA), 201 Insurance, 98 Intangible assets, 27 Integration of economies, 124 Interest-bearing eNaira, 194–195 Internal stakeholders, 248 International agency, assessing funding needed by, 205 International Energy Agency (IEA), 205 International Monetary Fund (IMF), 163, 175 International Standards Organization, 260 Internet of Things (IoT), 23, 84, 86, 255 Investment, 98 management, 175 target for electric vehicles infrastructure, 202 target for renewable energy generation, 201–202 target to set up smart cities, 203–204 IP Act (2000), 91 Keynesian model, 193 Know Your Customer (KYC), 106 Knowledge management cybersecurity, 40 Laptops, 40 Latent Dirichlet allocation (LDA), 179 Lender’s Club, 110 Lending policies, 205 Lending practices, 206 Link strength, 137 Logical controls, 35 Machine learning (ML), 86, 257 Malware assaults, 39 Management accounting, 148 Managerial discretion, 147 Managers and employees, 245–246 Market share, 102 Marketers, 218 Marketplace lending (MPL), 100 Massive data analysis approach, 59 Measurement model of data analysis, 224–226 Medal, 238 Mendeley database, 133 Mentor–Protege relationship, 250 Micro, small and medium enterprises (MSMEs), 208 Microsoft (Private companies), 91 Minerals, 30 Mission critical assets, 29 MIT Sloan School of Management, 87 Modern banking, 2 Money, 56 Money laundering, 60, 79, 124, 172–173 analysis of word cloud, 181 annual trend of publication, 61 citation network analysis, 67–69 co-authorship network analysis, 62 co-authorship network analysis as authors on, 62 co-authorship network analysis as countries, 64–66 co-occurrences network analysis on, 66–67 countries analysis, 62 data and methodology, 175–179 descriptive analysis, 179–180 historical background of, 58–60 impacts of money laundering on Indian economy, 126–127 investigating TF-IDF, 182–184 link between words, 181 literature review, 173–175 measurements to control money laundering terror funding in India, 125–126 nexus of, 125 relationship between money laundering and economic and financial performance in topic modelling framework, 184–185 research methodology and data, 60 results, 61 sources analysis, 61–62 Money lending, old concept of, 100 Mordor Intelligence, 89 Naïve Bayes classifiers, 258 Narcotic Drugs and Psychotropic Substances Act (NDPS), 59 NASSCOM, 89 National Association of Securities Dealer Automated Quotations, The, 98 National Strategy for Artificial Intelligence by NITI Aayog, 88 Natural catastrophes, 24, 34 Natural disasters, 34 Neo banks, 101 Net profit, hypothesis testing for, 16–17 Netnograhphy, 236 Nigeria CBDC, 190 current features of, 193–194 interest-bearing eNaira, 194–195 literature review, 191–193 no transaction costs, 195 redesigning, 194 security, 195–196 NIRAMAI, 88 NIST, 25 NITI Aayog, 88 Non-banking financial corporations (NBFCs), 100, 206 Non-performing Assets, 4 Non-performing loans (NPL), 162 Non-scheduled banks, 2, 4 Normalisation, 178 NPA, hypothesis testing for, 17 Offline P2P lending, 100 Olympic Games Tokyo (2020), 237 Olympics, 240 destination, 234 players, 239 Online P2P lending platforms, 100, 109–110 Online platforms, 40 Opaque system, 252 Open Government Data (OGD), 87 Opensource Stack software, 87 Oracle (Private companies), 91 Organisation-owned devices, 40 Organisational goals, 246 Organisations, 244 Organised retailers, 230 Organised retailing, 221, 223 Palantir Technologies (American software company), 87 Pandemic, 160 Parallel economy, 124 Partial least square (PLS), 224 Payment, 191 Payments Banks, 2, 4 Performance appraisal (PA), 244 Performance assessments, 249 Performance improvement, notifications of, 249–250 Performance management systems, 244–245 AI and SA, 252–257 AI and SA in PM systems, 257–262 components of, 245–246 formal processes, 249–250 informal processes, 250 inputs to PM systems, 250–251 issues in current PM systems, 251–252 monitoring, 246–247 outputs from PM systems, 251 planning, 246 principles of PM systems, 248–249 processes in PM systems, 249 reviewing, 247 rewarding, 247 stakeholders in PM systems, 247–248 tasks in PM Systems, 246 Person-to-Business lending models (P2B lending models), 101 Person-to-Person lending models (P2P lending models), 101, 109–111 crediting, 109 elements, 100 lending platforms, 100–101, 106 lending services, 104 Physical retailers, 216 Physical security current state of physical security controls in SME, 34 measures, 34 Physical stores, 217 Planning–programming–budgeting system, 84 Platform lending, 100 Policy-making process, 84 in India, 87 Predictive modelling techniques, 85 Prevention of Money Laundering Act (PMLA), 59, 126 Priority sector lending role of financial institutions through, 208–210 Priority theory of sustainable finance, 147–148 PRISMA 2020 Flow Diagram, 133 Private Banks, 8, 18 Private companies, 91 Private giants, 89 Private Sector Banks, 3 Production-linked Incentive Scheme (PLI Scheme), 207 Profitability, 5 of Indian banks, 7 Programmable integrated circuits (PLCs), 22 Public policies, 84 Public Sector Banks, 3, 8, 18 Python application, 112, 114 Qualitative research techniques, 85 Quality assessment, 134 Quasi-legal enterprises, 56 Real-time resource monitoring techniques, 85 Refugee Olympic, 237 Regional Rural Banks (RRB), 2–3 Relative frequency, 183 Renewable energy, 22, 201, 207 investment target for renewable energy generation, 201–202 Research methodology, 85 Reserve Bank of India (RBI), 2, 6, 99, 207 funds transfer mechanism as per RBI guidelines, 105 Reserve Bank of India Act (1934), 2, 4 Resource allocation, 76 Retail business analytics, 219 Retail customers, 218–219 Retail digital promotion, 220–221 Retail industry, 221 Retail scenario in India, 217–218 Retail stores, 216 Retailers, 216–217, 229 Retailing, 220 data analysis, 224–229 data collection, 223–224 development of conceptual model, 221–223 drivers of digitisation and AI in retail, 218–219 implications of study, 229–230 limitation, scope for further research, 230 literature review and hypothesis development, 217 research methodology, 223–223 research objectives, 217 retail digital promotion, 220–221 retail digitisation using AI, 219–220 retail scenario in India, 217–218 Reward system, 247 Rio Olympic (2016), 236 Scalability, 252 Schedule Banks, 4 Scheduled banks, 2 Search strategy, 133 Security awareness training for employees in SMEs, frequency of, 37–38 Security controls in SME, current state of, 34 Security objectives, 27, 29 Selection criterion, 133–134 Sentiment analysis, 238 method, 114 perspective, 111–112 Sentiments, 236 Sin stocks, 80 Small and Medium Enterprises (SME), 22, 31 age of, 32 analysis of research interview results, 41 analysis of research survey results, 32 biggest problems faced by SMEs implementing or deciding/planning to implement cybersecurity controls, 38–39 core cybersecurity concepts, 26–31 current state of administrative security controls in, 35–37 current state of implemented standards or framework in, 32–33 current state of physical security controls in, 34–35 current state of security controls in, 34 current state of technical security controls in, 35 DLCI framework, 41–48 DLCI Level, 45–46 DMCAS Level, 44 experience of cyberattacks faced by, 39–40 frequency of security awareness training for employees in, 37–38 literature review, 24–26 methodology, 31–32 minimum overall cybersecurity controls implementation level, calculate, 44–45 responsible AI for mapping cybersecurity controls for, 46–48 WCCI for, 44 Small and medium-sized businesses (SMBs), 23, 101 Small business owners, 109 Small finance banks, 2–4 Smart Analytics, 252, 254–255 ability of AI and SA to improve PM systems, 257 AI in HR management, 259 application of AI and SA in business processes, 255 automating PM Systems, 259–261 benefits of, 255 classification models to predict employee performance, 257–259 limitations of AI and SA in PM systems, 261–262 in PM systems, 257 SA in business processes, 256–257 Smart cities, investment target to set up, 203–204 Social accounting, 148 Social media, 219, 235 Social reporting, 146 Socially responsible investments (SRI), 74–76, 79–81 action plan for resolving financial crimes using SRI–ESG Model, 81 current market trend of SRI in India, 74–75 effects of financial crimes, 79 ESG and SRI, 79–81 findings of study, 77 heterogeneity, 76 in India, 75 investors, limitations of study, 77 research 77 research methodology, 77 review of socially responsible 75–76 76 terrorism financing devastating part of financial 78–79 impact of unethical investments on financial crimes, 79 unethical investors, 77–78 41 energy generation, 201 234 234 239 239 149 in PM Systems, 247–248 107 Bank of India 2 Data Analysis, 224 theory 235 modelling 224 model of data analysis, 226–229 178 chain management, 140 income model, implication of model, implications for or for sustainability, 151–152 profit, 206 149 Sustainable development, 206 Sustainable development Sustainable economic development, 140–141 Sustainable economic goals, role of financial institutions in 206–207 Sustainable 147 Sustainable growth in India, 206 security 26 18 249 (Private 89 98 controls, 35 security controls in SME, current state of, 35 132 and AI, 221 companies, 98 Technology and and model Technology governance data sciences in economic and social issues, 86–87 India’s stance on data-driven policy, 87–89 research and methodology, findings and 22 frequency 179 investigating TF-IDF, 182–184 financing on Indian economy, impacts of, 127–128 59 financing, 79 devastating part of financial 78–79 nexus of, 125 threat agents, 24 analysis, 178 data methods, 178 agents, 24 Tokyo Olympic (2020), 234 data analysis and findings, literature review, research methodology, 236 Tokyo Olympic 234 179 relationship between money laundering and economic and financial performance in, 184–185 profit, action 236 PM systems, 252 149 253 234 content analysis, 236 data, 236 sentiments, 235 laundering system, 59 investments on financial crimes, impact of, 79 investors, 77–78 payments 6 Union Bank, 7, 10 56 National Security 28 224 analysis, 60 60, 135 23 87 40 of 173 Cybersecurity Implementation 44 22 23 Yes Bank, 7, 18 25 COVID 19 and Financial Performance of Banks in and of Least Cybersecurity for Small and Medium Enterprises for Global Digitalised Economy Money of from to Socially for Financial Technology and of Governance and Policy-making for the the Sentiment Analysis of Cashkumar Peer-to-Peer Lending on Google of Money Laundering and Financing on the Indian Economy the on Implementation of in Earnings Management for Income of Sustainable COVID-19 Global Debt to the Analysis of Money Laundering Research the Central Bank Digital for Payments and Green Banking to Sustainable Economic Digitisation and Artificial Intelligence in Sector on Sentiment Analysis in 2020 Smart and AI for Modern Performance Management

Sentiment analysis, which stands for opinion mining, is a natural language processing (NLP) technique that involves identifying, extracting, and analyzing sentiments or opinions expressed in text data. The primary goal of sentiment analysis is to determine the sentiment polarity of a given piece of text, whether it is positive, negative, or neutral. This analysis can be applied to various types of content, such as product reviews, social media posts, customer feedback, and news articles. Sentiment analysis algorithms use machine learning and text classification to understand subjective information conveyed in text, helping businesses, organizations, and individuals gain insight into public opinions and emotions about specific topics, products, or services. In this study, we conducted sentiment analysis on a Bengali dataset. For feature extraction, we implemented the term frequency-inverse document frequency (TF-IDF) technique, and for feature selection, we applied an extra tree classifier approach. Subsequently, we trained our machine learning model, achieving an impressive accuracy rate of 92%.

This paper presents the performance of Twitter sentiment analysis on the Agricultural Ministry of India and also determines the optimal AI model between random forest (RF) and k nearest neighbours (KNN) algorithms. ‘Twitter and Reddit Sentimental Analysis’ dataset from Kaggle is used to train and test the RF and KNN AI models. Term frequency-inverse document frequency (TF-IDF) vectorizer is used to extract features from the obtained dataset. The optimal AI model is determined using the library functions of scikit-learn-2.0 which perform the training and testing of the dataset. Twitter authenticated API called Tweepy is used to create the dataset required to determine the public sentiment towards the Agriculture Ministry by us. This dataset includes 6000 tweets related to the Agriculture Ministry. The Twitter sentiment analysis performed determines what percentage of public opinion towards the Agriculture Ministry is positive, negative and neutral.KeywordsTwitter sentiment analysisRandom forest algorithmK nearest neighbours algorithmTweepyAgricultureTweets

Sentiment analysis is the interpretation and classification of emotions conveyed by text data. While there have been many attempts to classify the sentiment of a given text, there have been few models that can do the same when provided with non-English data exhibiting sarcasm or irony. This paper aims to compare various techniques of sarcasm detection and decide which method works the best for datasets of different sizes and types. The models have been tested on datasets of three different non-English languages - Arabic, French and a Hindi-English code-mix. None of the presented models are language-specific and can be run on data of any language. A comparison between a sub-word model, the usage of Term Frequency–Inverse Document Frequency (TF-IDF) and neural networks, a Long Short-Term Memory (LSTM) model and machine learning techniques such as Nearest Neighbors, Decision Tree, Random Forest, AdaBoost, Naive Bayes (NB), Support Vector Machine (SVM) Linear, SVM radial basis function (RBF), SVM Sigmoid has been performed. The output for each language and model has been evaluated based on their F1-score, accuracy, precision, and recall.

<p>In March of 2020, the World Health Organization identified COVID-19 as a new pandemic and issued a statement to that effect. This fatal virus was able to disperse and propagate throughout several countries all over the world. During the progression of the pandemic, social networking sites like Twitter generated significant and substantial volumes of data that helped improve the quality of decisions pertaining to health care applications. In this paper, we proposed a sentiment classification using various feature extraction and machine leavening techniques for social media dataset. The system has divided into four phase data collection, preprocessing and normalization, feature extraction and feature selection and finally classification. In first phase we collect data from social media sources such as twitter using Twitter API. In second phase the tweets, data was ready for preprocessing and it was sorted into three categories: positive, neutral, and negative. During the third phase, various features were extracted from the tweets by employing a number of widely utilized approaches, including as bag of words (BoW), Term Frequency-Inverse Document Frequency (TF-IDF), Word2Vec, and FastText, to gather feature datasets. These methods were employed to extract distinct datasets for the features. The final phase different machine learning classification algorithms are applied for detection of sentiment using machine learning. In the extensive experimental analysis, the BoW performed better results with modified support vector machine (mSVM) than existing machine learning algorithms. The proposed mSVM performed superiorly to the other classifiers by 98.15% accuracy rate. Once the tweets are correctly classified as COVID-19 tweets, it is further categorized into three sentiments that is positive, negative and neural. Proposed mSVM achieves 93% of accuracy rate for positive sentiment which better as compared to other Machine Learning (ML) classifiers.</p>

The goal of the Metadata Automation DREAM Challenge was to develop a tool to automate the annotation of metadata fields and values in structured biomedical data files with the best candidate Common Data Element (CDE) matches from the Cancer Data Standards Registry and Repository (caDSR). We chose to implement our model in Python 3.6 and approached this challenge from the perspective that it was essentially a fuzzy matching problem. Our approach utilizes Scikit-Learn’s TfidfVectorizer class along with a custom n-gram function to vectorize the data. These term frequency - inverse document frequency (TF-IDF) vectors are passed to Scikit-Learn’s Nearest Neighbor class which returns the k nearest CDE neighbors and their associated distance scores for each column header in the biomedical data file. For the returned CDEs with enumerated values, the Levenshtein distances from the observed values in the data to the CDE’s permissible values are computed using Python’s FuzzyWuzzy library. We then use a decision tree approach based on the TF-IDF distance scores and the observed values’ average Levenshtein distance scores to select and rank the top three CDE matches from the set of nearest neighbors for each column header. In this final ranking step, we apply cutoff values to the distance scores to determine when to include ‘NOMATCH’ as one of the three results. Throughout the challenge we experimented with many aspects of the algorithm including modifying the n-gram function, the selection of caDSR fields to include in the TF-IDF vectorization and applying different cutoff values. The final version of our model was arrived at by selecting the features and parameters that maximized the overall score across all the provided test datasets.

Sentiment analysis, a crucial aspect of Natural Language Processing (NLP), plays a pivotal role in understanding public opinion, customer feedback, and user sentiments in various domains. In this study, we present a comprehensive approach to sentiment analysis that incorporates both textual and emoji data, leveraging diverse datasets from sources such as social media, customer reviews, and surveys. Our methodology consists of several key steps, including data collection, pre-processing, feature extraction, feature fusion, and feature selection. For data pre-processing, we apply techniques such as tokenization, lowercasing, stop word removal, and stemming to ensure uniform and meaningful text representation. We also extract emojis from text using regular expressions and convert them into textual representations, facilitating unified processing. Text data is transformed into feature vectors using Term frequency_ Inverse Term frequency (TF-IDF) weighting-based Bag of Words (TF-IDF_BoW) with word frequencies, pre-trained Word2Vec word embeddings, and n-grams to capture local word patterns. Emoji data, on the other hand, is processed using pre-trained emoji embeddings (Emoji2Vec) and emoji frequency counting. The Feature Fusion stage involves the combination of text and emoji features into a single feature vector, with a weighted concatenation approach. In the Feature Selection phase, introduced a Self-Improved Siberian Tiger Optimization (SI-STO), a novel feature selection technique, to identify the most relevant features for sentiment prediction. The sentiment classification model SentEmoNet combines LSTM with an Attention Mechanism and a CNN for capturing text patterns and extracting local features from text and emoji embeddings. We use standard sentiment analysis metrics to assess performance in the model evaluation phase. The proposed model has achieved an accuracy of 97.87%, which is better than the existing ones.

Background: Text classification techniques are increasingly important with the exponential growth of textual data on the internet. Term Frequency-Inverse Document Frequency (TF-IDF) and Count Vectorizer(CV) are commonly used methods for feature extraction. TF-IDF assigning weights to terms based on their frequency. CV simply counts the occurrences of terms. The performance of CV as well as TF-IDF are evaluated and compared with KNN and DT classifiers across text datasets. Methodology: The investigation begins with preprocessing. The feature vectors are created using both TF-IDF and CV. Feature vectors are passed into the KNN and DT classifiers at in training stage. Experiments are executed the usage of Kaggle's public database Ukraine 10K tweets sentiment_analysis dataset and the Womens ecommerce clothing reviews dataset. Findings: The average of precision, recall, f1 score and accuracy of KNN with TF-IDF were 84.5%, 87%, 83%, 87% respectively and KNN with CV were 83.5%, 87%, 83.5%, 87% respectively. Similarly, average of precision, recall, f1 score and accuracy of DT with TF-IDF were 89%, 89%, 89%, 89% respectively and DT with CV were 89%, 89.5%, 89.5%, 89.5% respectively. The results obtained in this research is consistent with previous similar research result. Conclusions: The performance of TF-IDF is almost similar as CV for a particular dataset and a particular classifier in this study. Novelty: The experiment performed using these classifiers and feature extraction methods on the datasets is a novelty and contribution of this research.

Sentiment analysis is widely used today to make data-driven decisions in different industries, starting from marketing and including brand management, reputation monitoring, and customer satisfaction analysis. Its growing importance is closely linked with so-called ‘word-of-mouth’ communication, from reading online reviews to writing comments on social networks. Effective separation of sentiments ensures that companies' responses are timely and critical patterns are seen in big data sets. Statistical measures, information gain, correlation-based approaches, etc, have been employed for the feature selection. Still, the problem associated with text data mining is that they don’t convey the text's relative difficulty and additional features. To fill this gap, our research proposes a new feature selection technique through Ant Colony Optimization (ACO) and K Nearest Neighbour (KNN) performed on 28,000 customer reviews in different product categories. The results, therefore, showed an overall accuracy of 80.1%, with the Support Vector Machine (SVM) set at 80.5% on each selected feature, which was slightly higher than the Convolutional Neural Network (CNN), which scored a 78.41% accuracy. SVM remains on the mark of 83%, and for CNN, the rate achieved on the same was 80.8% when both were applied to the entire dataset. These facts rejected the infallibility of the simple and complex algorithms used singly in the sentiment classification, indicating that more sophisticated algorithms like ACO and KNN can provide business solutions to improve their service delivery based on customers’ feedback.

With the ever-increasing number of digital documents, the ability to automatically classify those documents both quickly and accurately is becoming more critical and difficult. A text classification system for Chinese documents is developed in this paper. A HTF-WDF algorithm is proposed for feature selection. Different from other feature selection algorithms, this method considers the effect of term frequency. Using the idea of fuzzy feature, the terms with high term frequency (HTF) are distinguished and appended to the feature list. The features which can represent the topic of the documents are picked out according to the weighted document frequencies (WDF), which can avoid the problems of the traditional document frequency (DF) method. Then the Support Vector Machine (SVM) is used to training the classifier. The proposed algorithm is verified by representative Chinese documents. The experiment results manifest the superiority of the proposed algorithm to the traditional DF algorithm.

Extreme learning approaches are used to perform sentiment analysis on restaurant evaluations. Sluggish training and overfitting are two problems that traditional supervised learning techniques frequently face. In order to prevent overfitting and speed up training, extreme learning makes use of single hidden layer neural networks with randomly assigned weights. The goal is to create a quick and accurate model for identifying sentiment in meal reviews and to assess the differences between supervised learning techniques and models based on extreme learning. In order to map the scores to attitudes, the study uses a dataset of food reviews, where scores larger than three are interpreted as favourable and scores below that as negative. Training and testing sets are created from the dataset following preparation, which includes handling missing values and choosing pertinent columns. For modeling purposes, text data is transformed into Term Frequency-Inverse Document Frequency (TF-IDF) characteristics. The network is given randomly initialized weights and biases in both single layer and multi layer perceptron implemented. During model training, the loss function is computed, weights and biases are adjusted by backpropagation, and predictions are computed using forward propagation technique. By using a threshold of 0.5, the model's accuracy is assessed. Accuracy scores are used as a statistic for reporting training and testing accuracy. The model further validates the effectiveness for sentiment analysis in the context of food reviews by showing quicker training times and less sensitivity to overfitting. The work presents extreme learning approaches as a competitive substitute for supervised learning, which advances sentiment analysis tools. Comparing the model based on extreme learning to traditional supervised learning methods, experimental results show that the latter achieves competitive accuracy in sentiment analysis. Faster training times and less sensitivity to overfitting are further features. As a strong substitute for supervised learning techniques, this study highlights the effectiveness of extreme learning approaches for sentiment analysis in meal evaluations. Extreme learning improves the efficacy and precision of sentiment analysis models, especially in areas such as restaurant evaluations, furthering the practical uses of sentiment analysis tools.

Our research developed a non signature based approach, employing feature selection methods such as Categorical Proportional Distance (CPD), Weight of Evidence of Text (WET), Term Frequency - Inverse Document Frequency (TF-IDF), Term Frequency - Inverse Document Frequency - Class Frequency (TF-IDF-CF), Galavotti-Sebastiani-Simi Coefficient (GSS) and Term Significance (TS). Classification model is developed by considering bi--gram features ranked with these feature selection techniques. The proposed feature selection approaches detect unseen malware samples with accuracy in the range of 99% to 100%. Relevance of a feature ranking methods on variable feature length is ascertained using McNemar test.

Sentiment analysis is a new area of research in data mining that concerns the detection of opinions and/or sentiments in texts. This work focuses on the application and the comparison of three classification techniques over a text corpus composed of reviews of commercial products in order to detect opinions about them. The chosen domain is about "perfumes", and user opinions composing the corpus are written in Italian language. The proposed approach is completely data-driven: a Term Frequency / Inverse Document Frequency (TFIDF) terms selection procedure has been applied in order to make computation more efficient, to improve the classification results and to manage some issues related to the specific classification procedure adopted.