Pub Date : 2023-02-11DOI: 10.1109/ICITIIT57246.2023.10068706
K. S. Sudheera, Swetha R, Tejaswini R, Vaishali Meena M, Anu G. Kumar
This paper presents residential load forecasting using multivariate multi-step Deep Neural Networks (DNN) such as LSTM, CNN, Stacked LSTM, and Hybrid CNN-LSTM. A preliminary Exploratory Data Analysis (EDA) is conducted, and the decision variables are identified. An elbowing method is used to determine the number of clusters. Data is categorized based on weekdays, weekends, vacations, and Covid-Lockdown. Dimensionality-reduction using principal component analysis (PCA) is conducted. Seasonality-based clustering is found to improve the DNN model prediction accuracy further. A comparative analysis employs error metrics such as RMSE, MSE, MAPE, and MAE. The multivariate LSTM model with feedback is found to be the best fit model with the better performance indices.
{"title":"Residential Load Forecasting based on Deep Neural Network","authors":"K. S. Sudheera, Swetha R, Tejaswini R, Vaishali Meena M, Anu G. Kumar","doi":"10.1109/ICITIIT57246.2023.10068706","DOIUrl":"https://doi.org/10.1109/ICITIIT57246.2023.10068706","url":null,"abstract":"This paper presents residential load forecasting using multivariate multi-step Deep Neural Networks (DNN) such as LSTM, CNN, Stacked LSTM, and Hybrid CNN-LSTM. A preliminary Exploratory Data Analysis (EDA) is conducted, and the decision variables are identified. An elbowing method is used to determine the number of clusters. Data is categorized based on weekdays, weekends, vacations, and Covid-Lockdown. Dimensionality-reduction using principal component analysis (PCA) is conducted. Seasonality-based clustering is found to improve the DNN model prediction accuracy further. A comparative analysis employs error metrics such as RMSE, MSE, MAPE, and MAE. The multivariate LSTM model with feedback is found to be the best fit model with the better performance indices.","PeriodicalId":170485,"journal":{"name":"2023 4th International Conference on Innovative Trends in Information Technology (ICITIIT)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121464244","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 : 2023-02-11DOI: 10.1109/ICITIIT57246.2023.10068647
N. Gayathri, Prawin R, Ranjith kumar A, M. R.
With the proliferation of virtual entertainment platforms around the world, particularly among young people, digital taunting and enmity have become real and annoying problems that networks must address. Threats can use these levels to attack and weaken others in their networks. To combat digital tormenting, various strategies and tactics have been used or proposed, including early detection and alarms that both detect and protect victims from such attacks. Machine Learning techniques with Artificial Intelligence Framework are being widely used to identify specific linguistic patterns that danger uses to hunt for their victims. The Feeling Analysis (FA) of virtual entertainment content is one of the expanding fields of research in AI. FA makes it possible to gradually identify online harassment and continuously recognize cyberbullying. This study recommends a SA model to identify cyberbullying messages in Facebook web-based entertainment. SVM and MAXENT classifier, controlled AI arrangement tools, are employed in this model. When a higher n-grams language model is applied to such texts in correlation with analogous prior research, the findings of the investigations carried out using this model showed encouraging results. Similar patterns in the results showed that these classifiers preferred execution measures over other classifiers on such remarks.
{"title":"Machine Learning techniques for identifying Cyberbullying on digital networks","authors":"N. Gayathri, Prawin R, Ranjith kumar A, M. R.","doi":"10.1109/ICITIIT57246.2023.10068647","DOIUrl":"https://doi.org/10.1109/ICITIIT57246.2023.10068647","url":null,"abstract":"With the proliferation of virtual entertainment platforms around the world, particularly among young people, digital taunting and enmity have become real and annoying problems that networks must address. Threats can use these levels to attack and weaken others in their networks. To combat digital tormenting, various strategies and tactics have been used or proposed, including early detection and alarms that both detect and protect victims from such attacks. Machine Learning techniques with Artificial Intelligence Framework are being widely used to identify specific linguistic patterns that danger uses to hunt for their victims. The Feeling Analysis (FA) of virtual entertainment content is one of the expanding fields of research in AI. FA makes it possible to gradually identify online harassment and continuously recognize cyberbullying. This study recommends a SA model to identify cyberbullying messages in Facebook web-based entertainment. SVM and MAXENT classifier, controlled AI arrangement tools, are employed in this model. When a higher n-grams language model is applied to such texts in correlation with analogous prior research, the findings of the investigations carried out using this model showed encouraging results. Similar patterns in the results showed that these classifiers preferred execution measures over other classifiers on such remarks.","PeriodicalId":170485,"journal":{"name":"2023 4th International Conference on Innovative Trends in Information Technology (ICITIIT)","volume":"14 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131968442","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 : 2023-02-11DOI: 10.1109/ICITIIT57246.2023.10068623
M. R., K. S. Nikhil
From the recent reported studies, it is clear that Ga2O3 can offer higher breakdown voltage due to its higher bandgap. However, Ga2O3 based power devices are having challenges like low carrier concentration and less electron mobility. In this article, a Junctionless Enhancement mode Field Effect Transistor (FET) with Ga2O3 REduced SURface Field (RESURF) is proposed. The introduction of n-type Ga2O3 RESURF region between gate and drain region improves the breakdown voltage. The asymmetric gate structure further enhances the breakdown voltage by delaying the attainment of critical electric field. The variation of on resistance (RON) for varying the length of RESURF region (Lr) is also investigated. Junctionless FET with Ga2O3 RESURF has shown large potential for high power integrated circuit applications.
{"title":"Improvement in Breakdown Voltage of Junctionless Power Transistor with Ga2O3 RESURF region","authors":"M. R., K. S. Nikhil","doi":"10.1109/ICITIIT57246.2023.10068623","DOIUrl":"https://doi.org/10.1109/ICITIIT57246.2023.10068623","url":null,"abstract":"From the recent reported studies, it is clear that Ga2O3 can offer higher breakdown voltage due to its higher bandgap. However, Ga2O3 based power devices are having challenges like low carrier concentration and less electron mobility. In this article, a Junctionless Enhancement mode Field Effect Transistor (FET) with Ga2O3 REduced SURface Field (RESURF) is proposed. The introduction of n-type Ga2O3 RESURF region between gate and drain region improves the breakdown voltage. The asymmetric gate structure further enhances the breakdown voltage by delaying the attainment of critical electric field. The variation of on resistance (RON) for varying the length of RESURF region (Lr) is also investigated. Junctionless FET with Ga2O3 RESURF has shown large potential for high power integrated circuit applications.","PeriodicalId":170485,"journal":{"name":"2023 4th International Conference on Innovative Trends in Information Technology (ICITIIT)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130968044","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 : 2023-02-11DOI: 10.1109/ICITIIT57246.2023.10068631
Premkumar Duraisamy, S. Yuvaraj, Yuvaraj Natarajan, V. Niranjani
In recent years the boom of internet and social media usage everyone spend their invaluable time in social media app and looking for the solution for all kind of their problems. This work analysis deeply on how recommendation system works and its types in different platforms. Most of the modern recommendation system use machine learning algorithms like linear regression, random forest regression and support vector model with collaborative filtering method. Recommendation is nothing but an choice making system. It is vary from person to person based on their interest, culture, locality, education background, interpersonal skills etc., The huge item can be filtered from one by one based on each parameter and finally it will reach the right recommendation item. The research community has worked tremendous way in the field of recommendation system and produced huge variety of result. This survey enlightening the ideas about variety of recommendation system and techniques used by the research community.
{"title":"An Overview of Different Types of Recommendations Systems - A Survey","authors":"Premkumar Duraisamy, S. Yuvaraj, Yuvaraj Natarajan, V. Niranjani","doi":"10.1109/ICITIIT57246.2023.10068631","DOIUrl":"https://doi.org/10.1109/ICITIIT57246.2023.10068631","url":null,"abstract":"In recent years the boom of internet and social media usage everyone spend their invaluable time in social media app and looking for the solution for all kind of their problems. This work analysis deeply on how recommendation system works and its types in different platforms. Most of the modern recommendation system use machine learning algorithms like linear regression, random forest regression and support vector model with collaborative filtering method. Recommendation is nothing but an choice making system. It is vary from person to person based on their interest, culture, locality, education background, interpersonal skills etc., The huge item can be filtered from one by one based on each parameter and finally it will reach the right recommendation item. The research community has worked tremendous way in the field of recommendation system and produced huge variety of result. This survey enlightening the ideas about variety of recommendation system and techniques used by the research community.","PeriodicalId":170485,"journal":{"name":"2023 4th International Conference on Innovative Trends in Information Technology (ICITIIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129551052","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 : 2023-02-11DOI: 10.1109/ICITIIT57246.2023.10068649
Snehal A. Lohi, Chinmay Bhatt
Crop disease detection has become an integral part of smart farming models. To perform this task, various intrusive & non-intrusive models are proposed by researchers. Intrusive models have higher deployment cost, higher complexity & contaminate underlying crops, due to which they are limited to clinical use cases. For non-intrusive methods, it is observed that most of these models are capable of achieving better performance under application-specific datasets, and cannot be scaled for larger datasets. To overcome this limitation, a novel crop disease detection & yield prediction model via multi-parametric bio-inspired feature representation is proposed in this text. The proposed model initially uses a crop-specific adaptive thresholding technique, which assists in efficient segmentation for different crop types. The segmented imagery is processed via multiple feature extraction units, which extract colour, shape, texture & convolutional features. These features are further processed via use of Genetic Algorithm (GA) based feature selection model, that implements feature variance maximization to identify optimal feature sets. The selected feature sets are classified using ensemble classification model that combines Support Vector Machines (SVMs), Multilayer Perceptron (MLP), Linear Regression (LR), Decision Tree (DT), and Naïve Bayes (NB) classifiers. These classifiers were selected based on their accuracy performance under different crop types. It was observed that SVM & LR had better performance for Soybean & Squash crops, MLP & LR had better performance for Potato & Pepper crops, while NB had better accuracy for Apple & Raspberry crops. Due to a combination of these adaptive classifiers, the proposed model is capable of achieving an accuracy of 99.5% across multiple datasets, which makes it highly useful for a wide variety of classification scenarios.
{"title":"Design of a Crop Disease Detection Model using Multi-parametric Bio-inspired Feature Representation and Ensemble Classification","authors":"Snehal A. Lohi, Chinmay Bhatt","doi":"10.1109/ICITIIT57246.2023.10068649","DOIUrl":"https://doi.org/10.1109/ICITIIT57246.2023.10068649","url":null,"abstract":"Crop disease detection has become an integral part of smart farming models. To perform this task, various intrusive & non-intrusive models are proposed by researchers. Intrusive models have higher deployment cost, higher complexity & contaminate underlying crops, due to which they are limited to clinical use cases. For non-intrusive methods, it is observed that most of these models are capable of achieving better performance under application-specific datasets, and cannot be scaled for larger datasets. To overcome this limitation, a novel crop disease detection & yield prediction model via multi-parametric bio-inspired feature representation is proposed in this text. The proposed model initially uses a crop-specific adaptive thresholding technique, which assists in efficient segmentation for different crop types. The segmented imagery is processed via multiple feature extraction units, which extract colour, shape, texture & convolutional features. These features are further processed via use of Genetic Algorithm (GA) based feature selection model, that implements feature variance maximization to identify optimal feature sets. The selected feature sets are classified using ensemble classification model that combines Support Vector Machines (SVMs), Multilayer Perceptron (MLP), Linear Regression (LR), Decision Tree (DT), and Naïve Bayes (NB) classifiers. These classifiers were selected based on their accuracy performance under different crop types. It was observed that SVM & LR had better performance for Soybean & Squash crops, MLP & LR had better performance for Potato & Pepper crops, while NB had better accuracy for Apple & Raspberry crops. Due to a combination of these adaptive classifiers, the proposed model is capable of achieving an accuracy of 99.5% across multiple datasets, which makes it highly useful for a wide variety of classification scenarios.","PeriodicalId":170485,"journal":{"name":"2023 4th International Conference on Innovative Trends in Information Technology (ICITIIT)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114367487","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 : 2023-02-11DOI: 10.1109/ICITIIT57246.2023.10068584
S. Raju, M. Srikanth, K. Guravaiah, P. Pandiyaan, B. Teja, K. S. Tarun
For last few years, there has been significant research on application of AI/ML algorithms in stock prediction and stock market. Prediction in stock market is challenging as it is affected by various factors related to global markets, domestic markets, company related and overall sentiments of people. Stock market prediction can be done based on three aspects that is fundamental analysis, technical analysis, and sentimental analysis. In this paper, we have reviewed various AI/ML algorithms that can be used in predicting stock markets. We have covered all the three aspects of prediction and AI/ML algorithms applied in eachone of them. After reviewing some research papers, we have implemented a model which has given us 85% accuracy, we have achieved 10.28% return from our model portfolio, in last three months and 175% return in last one year.
{"title":"A Three-Dimensional Approach for Stock Prediction Using AI/ML Algorithms: A Review & Comparison","authors":"S. Raju, M. Srikanth, K. Guravaiah, P. Pandiyaan, B. Teja, K. S. Tarun","doi":"10.1109/ICITIIT57246.2023.10068584","DOIUrl":"https://doi.org/10.1109/ICITIIT57246.2023.10068584","url":null,"abstract":"For last few years, there has been significant research on application of AI/ML algorithms in stock prediction and stock market. Prediction in stock market is challenging as it is affected by various factors related to global markets, domestic markets, company related and overall sentiments of people. Stock market prediction can be done based on three aspects that is fundamental analysis, technical analysis, and sentimental analysis. In this paper, we have reviewed various AI/ML algorithms that can be used in predicting stock markets. We have covered all the three aspects of prediction and AI/ML algorithms applied in eachone of them. After reviewing some research papers, we have implemented a model which has given us 85% accuracy, we have achieved 10.28% return from our model portfolio, in last three months and 175% return in last one year.","PeriodicalId":170485,"journal":{"name":"2023 4th International Conference on Innovative Trends in Information Technology (ICITIIT)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128917381","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}