Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037310
A. Waghumbare, Upasna Singh, Nihit Singhal
In recent years, Deep Convolutional Neural Networks (DCNNs) have demonstrated some promising results in classification of micro-Doppler (m-D) radar data in human activity recognition. Compared with camera-based, radar-based human activity recognition is robust to low light conditions, adverse weather conditions, long-range operations, through wall imaging etc. An indigenously developed “DIAT-J.1RADHAR” human activity recognition dataset comprising micro-Doppler signature images of six different activites like (i) person fight punching (boxing) during the one-to-one attack, (ii) person intruding for pre-attack surveillance (army marching), (iii) person training (army jogging), (iv) person shooting (or escaping) with a rifle (jumping with holding a gun), (v) stone/hand-grenade throwing for damage/blasting (stone-pelting/grenades-throwing), and (vi) person hidden translation for attack execution or escape (army crawling and compared performance of this data on various DCNN models. To reduce variations in data, we have cleaned data and make it suitable for DCNN model by using preprocessing methods such as re-scaling, rotation, width shift range, height shift range, sheer range, zoom range and horizontal flip etc. We used different DCNN pre-trained models such as VGG-16, VGG-19, and Inception V3. These models are fine-tuned and the resultant models are performing efficiently for human activity recognition in DIAT-μRadHAR human activity dataset.
{"title":"DCNN Based Human Activity Recognition Using Micro-Doppler Signatures","authors":"A. Waghumbare, Upasna Singh, Nihit Singhal","doi":"10.1109/IBSSC56953.2022.10037310","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037310","url":null,"abstract":"In recent years, Deep Convolutional Neural Networks (DCNNs) have demonstrated some promising results in classification of micro-Doppler (m-D) radar data in human activity recognition. Compared with camera-based, radar-based human activity recognition is robust to low light conditions, adverse weather conditions, long-range operations, through wall imaging etc. An indigenously developed “DIAT-J.1RADHAR” human activity recognition dataset comprising micro-Doppler signature images of six different activites like (i) person fight punching (boxing) during the one-to-one attack, (ii) person intruding for pre-attack surveillance (army marching), (iii) person training (army jogging), (iv) person shooting (or escaping) with a rifle (jumping with holding a gun), (v) stone/hand-grenade throwing for damage/blasting (stone-pelting/grenades-throwing), and (vi) person hidden translation for attack execution or escape (army crawling and compared performance of this data on various DCNN models. To reduce variations in data, we have cleaned data and make it suitable for DCNN model by using preprocessing methods such as re-scaling, rotation, width shift range, height shift range, sheer range, zoom range and horizontal flip etc. We used different DCNN pre-trained models such as VGG-16, VGG-19, and Inception V3. These models are fine-tuned and the resultant models are performing efficiently for human activity recognition in DIAT-μRadHAR human activity dataset.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"109 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128506502","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037445
Subhash Mondal, Ranjan Maity, Yash Raj Singh, Soumadip Ghosh, A. Nag
Coronary-Heart-Disease (CHD) risk increases daily due to the uncontrolled lifestyle of today's adult age group. The early detection of the disease can prevent unfortunate death due to heart-related complications. The Machine Learning (ML) technique is essential for the early diagnosis of CHD and for identifying its many contributing factor variables. To build the prediction model, we have used the dataset consisting of 4240 instances and 15 related features to predict the possibility of future risk of CHD in the next ten years. Initially, thirteen ML models were deployed with 10-fold cross-validation, reflecting the highest test accuracy of 91.28% for the Random Forest (RF) classifier. The models were turned further, and the boosting algorithms showed the highest accuracy of 91 % and above; the Gradient Boost (GB) classifier performed better with an accuracy of 92.11 %. The voting ensemble approaches using the best-performing boosting models, namely GB, HGB, XGB, CB, and LGBM, have been considered for the final prediction. The prediction results reflected an accuracy of 92.26%, an F1 score of 91.25%, a ROC-AUC score of 0.917, and the number of False Negatives (FN) values is about 6.25% of the total test dataset.
{"title":"Early Prediction of Coronary Heart Disease using Boosting-based Voting Ensemble Learning","authors":"Subhash Mondal, Ranjan Maity, Yash Raj Singh, Soumadip Ghosh, A. Nag","doi":"10.1109/IBSSC56953.2022.10037445","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037445","url":null,"abstract":"Coronary-Heart-Disease (CHD) risk increases daily due to the uncontrolled lifestyle of today's adult age group. The early detection of the disease can prevent unfortunate death due to heart-related complications. The Machine Learning (ML) technique is essential for the early diagnosis of CHD and for identifying its many contributing factor variables. To build the prediction model, we have used the dataset consisting of 4240 instances and 15 related features to predict the possibility of future risk of CHD in the next ten years. Initially, thirteen ML models were deployed with 10-fold cross-validation, reflecting the highest test accuracy of 91.28% for the Random Forest (RF) classifier. The models were turned further, and the boosting algorithms showed the highest accuracy of 91 % and above; the Gradient Boost (GB) classifier performed better with an accuracy of 92.11 %. The voting ensemble approaches using the best-performing boosting models, namely GB, HGB, XGB, CB, and LGBM, have been considered for the final prediction. The prediction results reflected an accuracy of 92.26%, an F1 score of 91.25%, a ROC-AUC score of 0.917, and the number of False Negatives (FN) values is about 6.25% of the total test dataset.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127385264","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037475
Asmi Choudhary, Avaneesh Kumar, R. Jain, Syed Abou Iltaf Hussain
Optimization is a group of mathematical strategies for resolving quantitative issues in a variety of fields. The industries are relentlessly working to optimize more than one objective which are often conflicting in nature. Hence researchers are shifting their focus towards the multi-objective optimization algorithm which computes a set of Non-dominated solutions (NDS) which predominates other solutions in the search space. Non-dominated Sorting Genetic Algorithm II (NSGA-II) is one such multi-objective optimization algorithm but it fails to compute an accurate result when applied to rocky datasets. In order to overcome the difficulties, we have integrated the Artificial Neural Network (ANN) and TOPSIS with NSGA-II. The ANN algorithm creates the objective functions and the TOPSIS algorithm creates a trade-off between the NDS for better exploration. For testing the applicability of our approach we have applied it for computing the machining parameters for turning Aluminum alloy 6061-T6 using a high speed steel tool so that the objective performances namely machining time, material removal rate (MRR) and surface roughness (SR) are optimized. For validating the approach two experiments are conducted at the optimized parameters and the parameters obtained by the traditional NSGA-II approach. The computed the relative error (RAE) between the simulated and the first experimental values which is 1.87% for machining time, 4.2% for MRR and 4.3% for SR and the simulated and the second experimental values which is 14.8% for machining time, 12% for MRR and 11.2% for SR. The RAE value is very less and within the acceptable limit for the result computed by the proposed approach. The strength of our proposed algorithm is its practical applicability and ability to provide an accurate solution to an industry problem and hence our model is suitable for industrial applications.
优化是解决各种领域定量问题的一组数学策略。各个行业都在不懈地努力优化多个目标，而这些目标往往在本质上是相互冲突的。因此，研究人员将重点转向多目标优化算法，该算法计算一组在搜索空间中占主导地位的非支配解(NDS)。非支配排序遗传算法II (non - dominant Sorting Genetic Algorithm II, NSGA-II)就是其中的一种多目标优化算法，但应用于岩石数据集时无法计算出准确的结果。为了克服这些困难，我们将人工神经网络(ANN)和TOPSIS集成到NSGA-II中。ANN算法创建目标函数，TOPSIS算法在NDS之间进行权衡，以便更好地进行探索。为了验证该方法的适用性，将其应用于6061-T6铝合金高速刀具车削加工参数的计算，优化了加工时间、材料去除率(MRR)和表面粗糙度(SR)。为了验证该方法，在优化参数和传统NSGA-II方法得到的参数下进行了两次实验。计算出模拟值与第一次实验值的相对误差(RAE)，加工时间为1.87%，MRR为4.2%，SR为4.3%;模拟值与第二次实验值的相对误差(RAE)，加工时间为14.8%，MRR为12%，SR为11.2%，RAE值很小，在可接受的范围内。我们提出的算法的优势在于它的实用性和为工业问题提供准确解决方案的能力，因此我们的模型适合工业应用。
{"title":"A Hybrid ANN coupled NTOPSIS Approach: An Intelligent Multi-Objective Framework for solving Engineering Problems","authors":"Asmi Choudhary, Avaneesh Kumar, R. Jain, Syed Abou Iltaf Hussain","doi":"10.1109/IBSSC56953.2022.10037475","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037475","url":null,"abstract":"Optimization is a group of mathematical strategies for resolving quantitative issues in a variety of fields. The industries are relentlessly working to optimize more than one objective which are often conflicting in nature. Hence researchers are shifting their focus towards the multi-objective optimization algorithm which computes a set of Non-dominated solutions (NDS) which predominates other solutions in the search space. Non-dominated Sorting Genetic Algorithm II (NSGA-II) is one such multi-objective optimization algorithm but it fails to compute an accurate result when applied to rocky datasets. In order to overcome the difficulties, we have integrated the Artificial Neural Network (ANN) and TOPSIS with NSGA-II. The ANN algorithm creates the objective functions and the TOPSIS algorithm creates a trade-off between the NDS for better exploration. For testing the applicability of our approach we have applied it for computing the machining parameters for turning Aluminum alloy 6061-T6 using a high speed steel tool so that the objective performances namely machining time, material removal rate (MRR) and surface roughness (SR) are optimized. For validating the approach two experiments are conducted at the optimized parameters and the parameters obtained by the traditional NSGA-II approach. The computed the relative error (RAE) between the simulated and the first experimental values which is 1.87% for machining time, 4.2% for MRR and 4.3% for SR and the simulated and the second experimental values which is 14.8% for machining time, 12% for MRR and 11.2% for SR. The RAE value is very less and within the acceptable limit for the result computed by the proposed approach. The strength of our proposed algorithm is its practical applicability and ability to provide an accurate solution to an industry problem and hence our model is suitable for industrial applications.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126584520","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037571
Javesh Dafda, Mansi Subhedar
In software defined networking, load balancing is a crucial management operation for moving traffic packets from source to destination. Ant Colony Optimization (ACO) was employed with dynamic load balancing to enhance SDN performance in existing works. In order to improve the search for the ideal path, response time, span-time, and energy consumption, it is proposed in this article to employ energy-aware routing with a Genetic Algorithm (GA) and ACO load balancing. The goals are to minimize energy consumption while maintaining a quality of service for user flows and to achieve link load balancing. Simulation results demonstrate that the proposed scheme performs better in terms of response time and energy consumption.
{"title":"Dynamic Load balancing in SDN using Energy Aware Routing and Optimization Algorithm","authors":"Javesh Dafda, Mansi Subhedar","doi":"10.1109/IBSSC56953.2022.10037571","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037571","url":null,"abstract":"In software defined networking, load balancing is a crucial management operation for moving traffic packets from source to destination. Ant Colony Optimization (ACO) was employed with dynamic load balancing to enhance SDN performance in existing works. In order to improve the search for the ideal path, response time, span-time, and energy consumption, it is proposed in this article to employ energy-aware routing with a Genetic Algorithm (GA) and ACO load balancing. The goals are to minimize energy consumption while maintaining a quality of service for user flows and to achieve link load balancing. Simulation results demonstrate that the proposed scheme performs better in terms of response time and energy consumption.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"252 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121881207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037459
Anika Kapoor, Shivani Goel
Anxiety disorders have seen an elevating number since the Covid-19 pandemic. This paper aims at identifying more about the various anxiety disorders using machine learning Techniques. Further, symptoms of the types of anxiety disorders: Generalized Anxiety Disorder, Panic Disorder, Post-Traumatic Stress Disorder, Obsessive-Compulsive Disorder and Social Anxiety Disorder are also discussed. The datasets used in the paper are collected by researchers from hospitals/organizations/educational institutions mainly through questionnaires and surveys. Some of the many Machine Learning techniques used for prediction of these anxiety disorders include Random Forest, Linear Regression, Support Vector Machine among others. Lastly, the performance metric for the techniques is presented here and henceforth, the result is drawn from this available data followed by the conclusion.
{"title":"Prediction of Anxiety Disorders using Machine Learning Techniques","authors":"Anika Kapoor, Shivani Goel","doi":"10.1109/IBSSC56953.2022.10037459","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037459","url":null,"abstract":"Anxiety disorders have seen an elevating number since the Covid-19 pandemic. This paper aims at identifying more about the various anxiety disorders using machine learning Techniques. Further, symptoms of the types of anxiety disorders: Generalized Anxiety Disorder, Panic Disorder, Post-Traumatic Stress Disorder, Obsessive-Compulsive Disorder and Social Anxiety Disorder are also discussed. The datasets used in the paper are collected by researchers from hospitals/organizations/educational institutions mainly through questionnaires and surveys. Some of the many Machine Learning techniques used for prediction of these anxiety disorders include Random Forest, Linear Regression, Support Vector Machine among others. Lastly, the performance metric for the techniques is presented here and henceforth, the result is drawn from this available data followed by the conclusion.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128091164","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037553
Priyam Porwal, M. Devare
The paper helps to predict the future citation value of a fresh dataset of research papers by considering the past values of the citation count of paper using univariate time series analysis models and evaluating their performance through various evaluation metrics. It is important to predict future citation count as it helps to assess researcher's achievements, promotions, fund allocation, etc. This research is in addition to past research where for prediction, different parameters like content of paper, author details, venue impact etc. were considered. The real and original data for the dataset was extracted from the Google Scholar profile of top ranked authors. Three models of time series, Autoregressive Integrated moving average(ARIMA), Simple exponential smoothing (SES), and Holt winter's exponential Smoothing (HWES) are applied to observe the result variations. The models obtained error metric values for the complete dataset. All four-evaluation metrics were calculated. The best results for the predictions for citation count were obtained from the Simple exponential smoothing and Holt winter's exponential Smoothing models, whose values were almost the same for all evaluation metrics because of almost no change in formula. Among all fourerror metrics mentioned in the design, MASE gave sensible results, with almost all values being less than 1. The results showed similar graphs for both Simple exponential smoothing and Holt winter's exponential smoothing models for actual and predicted values of citation count as there is negligible difference in formula.
{"title":"Citation Count Prediction Using Different Time Series Analysis Models","authors":"Priyam Porwal, M. Devare","doi":"10.1109/IBSSC56953.2022.10037553","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037553","url":null,"abstract":"The paper helps to predict the future citation value of a fresh dataset of research papers by considering the past values of the citation count of paper using univariate time series analysis models and evaluating their performance through various evaluation metrics. It is important to predict future citation count as it helps to assess researcher's achievements, promotions, fund allocation, etc. This research is in addition to past research where for prediction, different parameters like content of paper, author details, venue impact etc. were considered. The real and original data for the dataset was extracted from the Google Scholar profile of top ranked authors. Three models of time series, Autoregressive Integrated moving average(ARIMA), Simple exponential smoothing (SES), and Holt winter's exponential Smoothing (HWES) are applied to observe the result variations. The models obtained error metric values for the complete dataset. All four-evaluation metrics were calculated. The best results for the predictions for citation count were obtained from the Simple exponential smoothing and Holt winter's exponential Smoothing models, whose values were almost the same for all evaluation metrics because of almost no change in formula. Among all fourerror metrics mentioned in the design, MASE gave sensible results, with almost all values being less than 1. The results showed similar graphs for both Simple exponential smoothing and Holt winter's exponential smoothing models for actual and predicted values of citation count as there is negligible difference in formula.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114389496","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037443
Swapnil Singh, D. Krishnan, Pranit Sehgal, Harshit Sharma, Tarun Surani, Jay P. Singh
Sentiment Analysis has increasingly been used nowadays in many applications to evaluate opinion of public about products, policies, movies, politics. It is also used by government and law enforcement to understand behavior of people. One of the potential applications of sentiment analysis is candidate profiling and job recommendation. In the proposed research work, we evaluated the performance of supervised machine learning algorithms on dataset generated by us from twitter and indeed. We illustrated the steps involved in preproccesing the dataset generated through web scraping and making it ready for feeding into supervised algorithms. From our experimental study it is observed that Gradient Boosting Classifier gave the highest classification accuracy of 78.08 percent and AUC score of 0.819 on the test dataset.
{"title":"Gradient Boosting Approach for Sentiment Analysis for Job Recommendation and Candidate Profiling","authors":"Swapnil Singh, D. Krishnan, Pranit Sehgal, Harshit Sharma, Tarun Surani, Jay P. Singh","doi":"10.1109/IBSSC56953.2022.10037443","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037443","url":null,"abstract":"Sentiment Analysis has increasingly been used nowadays in many applications to evaluate opinion of public about products, policies, movies, politics. It is also used by government and law enforcement to understand behavior of people. One of the potential applications of sentiment analysis is candidate profiling and job recommendation. In the proposed research work, we evaluated the performance of supervised machine learning algorithms on dataset generated by us from twitter and indeed. We illustrated the steps involved in preproccesing the dataset generated through web scraping and making it ready for feeding into supervised algorithms. From our experimental study it is observed that Gradient Boosting Classifier gave the highest classification accuracy of 78.08 percent and AUC score of 0.819 on the test dataset.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123238116","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037560
Vandana Bharadi, Satya Prakash Awasthi
Student's performance analysis has taken a leap of faith in past two years when the delivery mode was shuttling between online and offline. Various factors which are significantly affecting student's performance are now newly to be researched and identified. Its very important to not only consider and study the effect of various academic factors but also socio-economic factors are needed to analyzed. Predictive analytics has shown its capabilities in efficiently predicting results in wide areas of application including academics. This analysis and prediction is most crucial in the developing country like India, where the published rate of retention of students at university level considered very low. In this research, the academic and socio-economic details collected from student through survey. Further efficacy of various machine-learning algorithms assessed by running these algorithms on survey data. The findings demonstrate that some machine learning algorithms may create accurate predictive models using historical data on student retention.
{"title":"Variables identification for Students Performance Prediction","authors":"Vandana Bharadi, Satya Prakash Awasthi","doi":"10.1109/IBSSC56953.2022.10037560","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037560","url":null,"abstract":"Student's performance analysis has taken a leap of faith in past two years when the delivery mode was shuttling between online and offline. Various factors which are significantly affecting student's performance are now newly to be researched and identified. Its very important to not only consider and study the effect of various academic factors but also socio-economic factors are needed to analyzed. Predictive analytics has shown its capabilities in efficiently predicting results in wide areas of application including academics. This analysis and prediction is most crucial in the developing country like India, where the published rate of retention of students at university level considered very low. In this research, the academic and socio-economic details collected from student through survey. Further efficacy of various machine-learning algorithms assessed by running these algorithms on survey data. The findings demonstrate that some machine learning algorithms may create accurate predictive models using historical data on student retention.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126433665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037276
Pratik. B. Kamble, B. Jadhav
The fermentation process is one of the most important processes in wine making. A certain amount of ratio of chemical compounds in red wine grape juice provides good quality wine. Acidifying and deacidifying grapes juice process is very complicated and non-linear, and ambiguous. Before starting the fermentation process the optimum balance of acid and pH is necessary. The purpose of this study is to develop a fuzzy expert system, this system can easily manipulate how much amount of acid or carbonates are required in red wine grape juice which saves time and gives good quality to the wine. A fuzzy interference system is used, if the acid level is low i.e. below 5 g/L then the acidification process will be carried out if the acid level is high i.e. above 8 g/L deacidification process will be carried out. A fuzzy rule base system handles uncertainty and gives a decision on acidifying and deacidifying processes. Domain expert takes trials of tartaric acid and pH values to get the optimum required amount of tartaric acid and carbonates value which is a time-consuming task. According to results, this system can easily manipulate how much amount of acid or carbonates are required in red wine grape juice which saves time and gives good quality to the wine.
{"title":"Fuzzy Expert System for Acidification and Deacidification Process in Red Wine Grape Juice","authors":"Pratik. B. Kamble, B. Jadhav","doi":"10.1109/IBSSC56953.2022.10037276","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037276","url":null,"abstract":"The fermentation process is one of the most important processes in wine making. A certain amount of ratio of chemical compounds in red wine grape juice provides good quality wine. Acidifying and deacidifying grapes juice process is very complicated and non-linear, and ambiguous. Before starting the fermentation process the optimum balance of acid and pH is necessary. The purpose of this study is to develop a fuzzy expert system, this system can easily manipulate how much amount of acid or carbonates are required in red wine grape juice which saves time and gives good quality to the wine. A fuzzy interference system is used, if the acid level is low i.e. below 5 g/L then the acidification process will be carried out if the acid level is high i.e. above 8 g/L deacidification process will be carried out. A fuzzy rule base system handles uncertainty and gives a decision on acidifying and deacidifying processes. Domain expert takes trials of tartaric acid and pH values to get the optimum required amount of tartaric acid and carbonates value which is a time-consuming task. According to results, this system can easily manipulate how much amount of acid or carbonates are required in red wine grape juice which saves time and gives good quality to the wine.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"2014 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127436191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-08DOI: 10.1109/IBSSC56953.2022.10037362
A. Manekar, Lochan Jolly
EEG signals convey information about a person's mental state, such as brain activity or degree of consciousness. Alcohol can also influence a person's degree of alertness. Long-term alcohol usage can cause certain patterns in EEG signals to emerge. Manual EEG signal analysis approach is difficult and time deterrent. As a result, neurologists make use of automated techniques to evaluate EEG data from their frequency sub-bands. The two separate brain states, alcoholism and normal, are identified in the current work utilizing Discrete Wavelet Transform technique for feature extraction from electroencephalogram (EEG) recordings. From the EEG signals under analysis, the sub-band coefficients using wavelet decomposition using Daubechies 7 basis wavelets are calculated. From the selected wavelet coefficients, statistical parameters including Minimum, Maximum, Average, Kurtosis, Mean square, and Standard-deviation are retrieved. In this research, this data is then sent to classifiers like Ensemble boosted trees, SVM, neural networks, and decision trees to distinguish between alcoholic and non-alcoholic EEG signals. While calculating accuracy ten-fold cross-validation is used to train the data. We discovered that the best results were provided by Ensemble boosted trees, with an Accuracy of 95.6 percent, Sensitivity of 91.3 percent, and FI score of 95.5 percent.
{"title":"Wavelet Decomposition based Automated Alcoholism Classification using EEG Signal","authors":"A. Manekar, Lochan Jolly","doi":"10.1109/IBSSC56953.2022.10037362","DOIUrl":"https://doi.org/10.1109/IBSSC56953.2022.10037362","url":null,"abstract":"EEG signals convey information about a person's mental state, such as brain activity or degree of consciousness. Alcohol can also influence a person's degree of alertness. Long-term alcohol usage can cause certain patterns in EEG signals to emerge. Manual EEG signal analysis approach is difficult and time deterrent. As a result, neurologists make use of automated techniques to evaluate EEG data from their frequency sub-bands. The two separate brain states, alcoholism and normal, are identified in the current work utilizing Discrete Wavelet Transform technique for feature extraction from electroencephalogram (EEG) recordings. From the EEG signals under analysis, the sub-band coefficients using wavelet decomposition using Daubechies 7 basis wavelets are calculated. From the selected wavelet coefficients, statistical parameters including Minimum, Maximum, Average, Kurtosis, Mean square, and Standard-deviation are retrieved. In this research, this data is then sent to classifiers like Ensemble boosted trees, SVM, neural networks, and decision trees to distinguish between alcoholic and non-alcoholic EEG signals. While calculating accuracy ten-fold cross-validation is used to train the data. We discovered that the best results were provided by Ensemble boosted trees, with an Accuracy of 95.6 percent, Sensitivity of 91.3 percent, and FI score of 95.5 percent.","PeriodicalId":426897,"journal":{"name":"2022 IEEE Bombay Section Signature Conference (IBSSC)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124438400","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：