Pub Date : 2022-12-17DOI: 10.1109/ICCIT57492.2022.10055515
R. Chowdhury, Farhad Uz Zaman, Arman Sharker, Mashfiq Rahman, F. Shah
In this contemporary era of digital marketing, ecommerce has emerged as one of the most preferred methods for day-to-day shopping. Ever since the COVID-19 pandemic, online shopping behavior has forever changed to less or no human-to-human interaction. As a result, it is getting more difficult for e-commerce enterprises to observe and evaluate market trends, particularly when done through consumer behavior analysis. To identify behavioral patterns and customer review-rating discrepancies, extensive analysis of product reviews is a substantial research field. Lack of benchmark corpora and language processing techniques, predicting review ratings in Bengali has become increasingly problematic. This paper thoroughly analyzes the approach to product review rating prediction for Bengali text reviews exploiting our own constructed dataset that was collected from an e-commerce website called DarazBD1. We acquired product reviews with labels known as ratings of five sentiment classes, from "1" to "5". It is noteworthy that we established a well-balanced dataset using our automated scraping system and a significant amount of time and effort is spent to maintain quality standards through the human annotation process. Exploration of multiple approaches to machine learning models such as logistic regression, random forest, multinomial naïve Bayes, and support vector machine, the best classification accuracy score of 78.63% is achieved by SVM. Subsequently, using Word2Vec, FastText, and GloVe embeddings with three deep neural network(DNN) architectures: CNN, Bi-LSTM, and a combination of CNN and Bi-LSTM, CNN+Bi-LSTM gave the highest accuracy score of 75.25% among the DNN architectures.
{"title":"Rate Insight: A Comparative Study on Different Machine Learning and Deep Learning Approaches for Product Review Rating Prediction in Bengali Language","authors":"R. Chowdhury, Farhad Uz Zaman, Arman Sharker, Mashfiq Rahman, F. Shah","doi":"10.1109/ICCIT57492.2022.10055515","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10055515","url":null,"abstract":"In this contemporary era of digital marketing, ecommerce has emerged as one of the most preferred methods for day-to-day shopping. Ever since the COVID-19 pandemic, online shopping behavior has forever changed to less or no human-to-human interaction. As a result, it is getting more difficult for e-commerce enterprises to observe and evaluate market trends, particularly when done through consumer behavior analysis. To identify behavioral patterns and customer review-rating discrepancies, extensive analysis of product reviews is a substantial research field. Lack of benchmark corpora and language processing techniques, predicting review ratings in Bengali has become increasingly problematic. This paper thoroughly analyzes the approach to product review rating prediction for Bengali text reviews exploiting our own constructed dataset that was collected from an e-commerce website called DarazBD1. We acquired product reviews with labels known as ratings of five sentiment classes, from \"1\" to \"5\". It is noteworthy that we established a well-balanced dataset using our automated scraping system and a significant amount of time and effort is spent to maintain quality standards through the human annotation process. Exploration of multiple approaches to machine learning models such as logistic regression, random forest, multinomial naïve Bayes, and support vector machine, the best classification accuracy score of 78.63% is achieved by SVM. Subsequently, using Word2Vec, FastText, and GloVe embeddings with three deep neural network(DNN) architectures: CNN, Bi-LSTM, and a combination of CNN and Bi-LSTM, CNN+Bi-LSTM gave the highest accuracy score of 75.25% among the DNN architectures.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132033898","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 : 2022-12-17DOI: 10.1109/ICCIT57492.2022.10056008
K. Hasan, Md. Sakhawat Hossain, Md. Abdulla Al-Sun, Md. Mostafizur Rahman
We describe a new method of parsing Bangla language based on Deterministic Finite Automata (DFA) and implement the parser using a TRIE data structure. Hence we call the parser as TRIE parser. TRIE parser successfully parses sentences faster than other important parsing schemes as it needs no formal rules, no parameters and no Context Free Grammars (CFG). the scheme stores the Bangla grammar symbols or Pasts Of Speech (POS) as a state of the DFA and process a sentence following the operations of a DFA. If the set of POS symbols reaches to final state, then parsing is successful otherwise unsuccessful. The parser uses the grammar rules in compressed form hence it becomes very less space consuming. Therefore, it can be implemented in light weight fashion in main memory. The TRIE parser is compared with two other parsers and it shows that the proposed TRIE parser outperforms others in terms of processing time with an increasing number of sentences in the input paragraph. Necessary figures and examples are used to properly explain the proposed TRIE parser.
{"title":"Designing a Bangla Parser using TRIE Based on Deterministic Finite Automata","authors":"K. Hasan, Md. Sakhawat Hossain, Md. Abdulla Al-Sun, Md. Mostafizur Rahman","doi":"10.1109/ICCIT57492.2022.10056008","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10056008","url":null,"abstract":"We describe a new method of parsing Bangla language based on Deterministic Finite Automata (DFA) and implement the parser using a TRIE data structure. Hence we call the parser as TRIE parser. TRIE parser successfully parses sentences faster than other important parsing schemes as it needs no formal rules, no parameters and no Context Free Grammars (CFG). the scheme stores the Bangla grammar symbols or Pasts Of Speech (POS) as a state of the DFA and process a sentence following the operations of a DFA. If the set of POS symbols reaches to final state, then parsing is successful otherwise unsuccessful. The parser uses the grammar rules in compressed form hence it becomes very less space consuming. Therefore, it can be implemented in light weight fashion in main memory. The TRIE parser is compared with two other parsers and it shows that the proposed TRIE parser outperforms others in terms of processing time with an increasing number of sentences in the input paragraph. Necessary figures and examples are used to properly explain the proposed TRIE parser.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132214725","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 : 2022-12-17DOI: 10.1109/ICCIT57492.2022.10055705
N. Sultana, R. Sultana, Risul Islam Rasel, M. M. Hoque
Low-resource natural language processing is getting more attention nowadays. Aspect-Based Sentiment Analysis (ABSA) from a high-resource language such as English becomes unchallenging because of sufficient datasets and experimentation tools. However, Aspect-Based Sentiment Analysis from low-resource languages such as Bangla is quite hard. So, many researchers are investing their time and knowledge in low-resource natural language processing. In this paper, we are proposing a Bangla Aspect-Based Sentiment Analysis model using Bangla natural language processing. We have collected 4012 Bangla text comments related to cricket, drama, movie, and music from YouTube. We have applied some very prominent supervised machine learning techniques such as Support Vector Classifier (SVC), Random Forest (RF), and Linear Regression (LR). We have achieved more than 75% accuracy in classifying positive, negative, and neutral sentiments and 80% accuracy in extracting aspects from Bangla texts. Finally, we used publicly available datasets to test our proposed model's generalizability. Furthermore, we find that our proposed approach surpasses earlier related research.
{"title":"Aspect-Based Sentiment Analysis of Bangla Comments on Entertainment Domain","authors":"N. Sultana, R. Sultana, Risul Islam Rasel, M. M. Hoque","doi":"10.1109/ICCIT57492.2022.10055705","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10055705","url":null,"abstract":"Low-resource natural language processing is getting more attention nowadays. Aspect-Based Sentiment Analysis (ABSA) from a high-resource language such as English becomes unchallenging because of sufficient datasets and experimentation tools. However, Aspect-Based Sentiment Analysis from low-resource languages such as Bangla is quite hard. So, many researchers are investing their time and knowledge in low-resource natural language processing. In this paper, we are proposing a Bangla Aspect-Based Sentiment Analysis model using Bangla natural language processing. We have collected 4012 Bangla text comments related to cricket, drama, movie, and music from YouTube. We have applied some very prominent supervised machine learning techniques such as Support Vector Classifier (SVC), Random Forest (RF), and Linear Regression (LR). We have achieved more than 75% accuracy in classifying positive, negative, and neutral sentiments and 80% accuracy in extracting aspects from Bangla texts. Finally, we used publicly available datasets to test our proposed model's generalizability. Furthermore, we find that our proposed approach surpasses earlier related research.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131022975","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 : 2022-12-17DOI: 10.1109/ICCIT57492.2022.10055757
Chowdhury Mohammad Abdullah, M. Kamal, Fairuz Shaiara, A. Kamal, Md. Azam Hossain
Blood transfusion is an integral part of the healthcare system that plays an important role in ensuring the quality of care for patients undergoing a variety of medical procedures and treatments. A large portion of this blood comes from voluntary donors. The existing blood donor management systems are unable to offer a reliable audit trail and traceability. Hence, there is a significant risk that patients may get transfusion of blood from unreliable sources. In this paper, we propose a system built on Ethereum with the goal of creating a decentralized, transparent, traceable, and secure network of blood donors. The platform uses smart contracts to facilitate peer-to-peer interactions. To encourage donors to donate blood more regularly, the system also offers rewards in the form of tokens. Our source code is available in a public Github repository1.
{"title":"BloodComm: A Peer-to-Peer Blockchain-based Community for Blood Donation Network","authors":"Chowdhury Mohammad Abdullah, M. Kamal, Fairuz Shaiara, A. Kamal, Md. Azam Hossain","doi":"10.1109/ICCIT57492.2022.10055757","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10055757","url":null,"abstract":"Blood transfusion is an integral part of the healthcare system that plays an important role in ensuring the quality of care for patients undergoing a variety of medical procedures and treatments. A large portion of this blood comes from voluntary donors. The existing blood donor management systems are unable to offer a reliable audit trail and traceability. Hence, there is a significant risk that patients may get transfusion of blood from unreliable sources. In this paper, we propose a system built on Ethereum with the goal of creating a decentralized, transparent, traceable, and secure network of blood donors. The platform uses smart contracts to facilitate peer-to-peer interactions. To encourage donors to donate blood more regularly, the system also offers rewards in the form of tokens. Our source code is available in a public Github repository1.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"149 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133048961","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 : 2022-12-17DOI: 10.1109/ICCIT57492.2022.10055592
Risul Islam Rasel, Anower Hossen Zihad, N. Sultana, M. M. Hoque
News Categorization is one of the primary applications of Text Classification, especially, Fake news classification. In recent days, many researchers have done plenty of work on Fake news detection in rich resource languages like English. But, due to a lack of resources and language processing tools, research on low-resource languages like Bangla is still insignificant. In this study, we try to build a Bangla Fake news dataset combining newly collected fake news data and available secondary datasets. Previously available datasets contained redundant data, which we reduced in our experiment. Finally, we build a Fake news dataset that contains 4678 distinct news data. We experimented with our data with multiple Machine Learning (LR, SVM, KNN, MNB, Adaboost, and DT), Deep Neural Networks (LSTM, BiLSTM, CNN, LSTM-CNN, BiLSTM-CNN), and Transformer (Bangla-BERT, m-BERT) models to attain some state of the art results. The best performing models are CNN, CNN-LSTM, and BiLSTM, with the accuracy of 95.9%, 95.5%, and 95.3%, respectively. We also tested our models by applying the previously existing datasets, and we got a 1.4% to 3.4% improvement in accuracy from previous results. Besides accuracy improvement, our models show a significant increase in recall of fake news data compared to the prior studies.
{"title":"Bangla Fake News Detection using Machine Learning, Deep Learning and Transformer Models","authors":"Risul Islam Rasel, Anower Hossen Zihad, N. Sultana, M. M. Hoque","doi":"10.1109/ICCIT57492.2022.10055592","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10055592","url":null,"abstract":"News Categorization is one of the primary applications of Text Classification, especially, Fake news classification. In recent days, many researchers have done plenty of work on Fake news detection in rich resource languages like English. But, due to a lack of resources and language processing tools, research on low-resource languages like Bangla is still insignificant. In this study, we try to build a Bangla Fake news dataset combining newly collected fake news data and available secondary datasets. Previously available datasets contained redundant data, which we reduced in our experiment. Finally, we build a Fake news dataset that contains 4678 distinct news data. We experimented with our data with multiple Machine Learning (LR, SVM, KNN, MNB, Adaboost, and DT), Deep Neural Networks (LSTM, BiLSTM, CNN, LSTM-CNN, BiLSTM-CNN), and Transformer (Bangla-BERT, m-BERT) models to attain some state of the art results. The best performing models are CNN, CNN-LSTM, and BiLSTM, with the accuracy of 95.9%, 95.5%, and 95.3%, respectively. We also tested our models by applying the previously existing datasets, and we got a 1.4% to 3.4% improvement in accuracy from previous results. Besides accuracy improvement, our models show a significant increase in recall of fake news data compared to the prior studies.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133158581","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 : 2022-12-17DOI: 10.1109/ICCIT57492.2022.10055219
Ishmamur Rahman, M. K. Islam, Abu Nowshed Chy, Muhammad Anwarul Azim
A skin lesion is an unusual change of skin tissues. While this can be caused by harmless skin diseases, there is also the chance of the lesion being cancerous. Skin cancer is one of the most common and deadly cancers in the world, which is caused by exposure to the ultraviolet radiation emitted by the sun. Due to the difficulty in visually differentiating between harmless and cancerous skin lesions, people are less likely to get medical attention straight away. Early diagnosis is crucial to ensure an effective treatment. Clinical and dermoscopy based diagnosis of cancerous skin lesions is costly, painful and sometimes inaccurate. Various researches report performing the classification of skin lesions using image processing techniques. Previous works in this domain are plenty, which reported fairly good results, where image processing and the use of both machine learning and deep learning models are seen. In this research, we propose a novel method which focused on important feature extraction, and fusing multiple features to improve the classification of malignant skin cells using traditional machine learning models, despite having imbalanced data distribution. The ISIC 2018 challenge dataset HAM10000 was used in our work. After preprocessing, we extracted shallow and deep features from the images. Shallow features consisted of position-wise color features and Scale Invariant Feature Transform (SIFT) features. Deep features were extracted by a transfer learning model MobileNetV3, which is pre-trained on Imagenet. These features were combined to form a more representative feature for the data. We parameter tuned five machine learning classifiers to do a binary classification on the processed data. The best accuracy, 81%, was obtained by using Support Vector Machine with an f1-score of 68%. Second best results were achieved by Random Forest Classifier, with an accuracy and F1-score of 80% and 67% respectively.
{"title":"Fusion of Shallow and Deep Features for Classifying Skin Lesions","authors":"Ishmamur Rahman, M. K. Islam, Abu Nowshed Chy, Muhammad Anwarul Azim","doi":"10.1109/ICCIT57492.2022.10055219","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10055219","url":null,"abstract":"A skin lesion is an unusual change of skin tissues. While this can be caused by harmless skin diseases, there is also the chance of the lesion being cancerous. Skin cancer is one of the most common and deadly cancers in the world, which is caused by exposure to the ultraviolet radiation emitted by the sun. Due to the difficulty in visually differentiating between harmless and cancerous skin lesions, people are less likely to get medical attention straight away. Early diagnosis is crucial to ensure an effective treatment. Clinical and dermoscopy based diagnosis of cancerous skin lesions is costly, painful and sometimes inaccurate. Various researches report performing the classification of skin lesions using image processing techniques. Previous works in this domain are plenty, which reported fairly good results, where image processing and the use of both machine learning and deep learning models are seen. In this research, we propose a novel method which focused on important feature extraction, and fusing multiple features to improve the classification of malignant skin cells using traditional machine learning models, despite having imbalanced data distribution. The ISIC 2018 challenge dataset HAM10000 was used in our work. After preprocessing, we extracted shallow and deep features from the images. Shallow features consisted of position-wise color features and Scale Invariant Feature Transform (SIFT) features. Deep features were extracted by a transfer learning model MobileNetV3, which is pre-trained on Imagenet. These features were combined to form a more representative feature for the data. We parameter tuned five machine learning classifiers to do a binary classification on the processed data. The best accuracy, 81%, was obtained by using Support Vector Machine with an f1-score of 68%. Second best results were achieved by Random Forest Classifier, with an accuracy and F1-score of 80% and 67% respectively.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132565505","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 : 2022-12-17DOI: 10.1109/ICCIT57492.2022.10054703
Aroni Saha Prapty, K. Hasan
Natural language processing is referred to as NLP that applies computational techniques for inter-communication between human and computer through human natural language on the basis of computer science, computational linguistic and artificial intelligence. The progression of NLP in different revolutionary techniques, word embedding has brought magnificent changes in the field of computational linguistic, statistical inference and so on. Semantic clustering can be interpreted as classify the group of identical objects that are semantically analogous. The main focus of the work is to manifest different word embedding techniques for semantic clustering of natural Bangla words. Earlier N-gram models were applied for the relevant field but dynamic word clustering models are currently popular due to the advancement of deep learning techniques because they speed up memory retrieval and decrease processing time. We discuss the effectiveness of Word2Vec, TF-IDF, FastText and GloVe word embedding models in this work and appraise the performance based on the models accuracy and competence.
自然语言处理(Natural language processing，简称NLP)是在计算机科学、计算语言学和人工智能的基础上，运用计算技术，通过人类的自然语言实现人与计算机之间的相互通信。自然语言处理在不同革命性技术中的发展，如词嵌入，给计算语言学、统计推理等领域带来了巨大的变化。语义聚类可以解释为对语义相似的相同对象组进行分类。本文的研究重点是揭示不同的词嵌入技术对自然孟加拉语词语义聚类的影响。早期的N-gram模型被应用于相关领域，但由于深度学习技术的进步，动态词聚类模型目前很流行，因为它们加快了记忆检索和减少了处理时间。本文讨论了Word2Vec、TF-IDF、FastText和GloVe四种词嵌入模型的有效性，并从模型的准确性和能力两方面对模型的性能进行了评价。
{"title":"Semantic Clustering of Bangla Natural Word using Different Word Embedding Techniques","authors":"Aroni Saha Prapty, K. Hasan","doi":"10.1109/ICCIT57492.2022.10054703","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10054703","url":null,"abstract":"Natural language processing is referred to as NLP that applies computational techniques for inter-communication between human and computer through human natural language on the basis of computer science, computational linguistic and artificial intelligence. The progression of NLP in different revolutionary techniques, word embedding has brought magnificent changes in the field of computational linguistic, statistical inference and so on. Semantic clustering can be interpreted as classify the group of identical objects that are semantically analogous. The main focus of the work is to manifest different word embedding techniques for semantic clustering of natural Bangla words. Earlier N-gram models were applied for the relevant field but dynamic word clustering models are currently popular due to the advancement of deep learning techniques because they speed up memory retrieval and decrease processing time. We discuss the effectiveness of Word2Vec, TF-IDF, FastText and GloVe word embedding models in this work and appraise the performance based on the models accuracy and competence.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133210277","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 : 2022-12-17DOI: 10.1109/ICCIT57492.2022.10054697
Dipannita Biswas, K. M. Azharul Hasan, Zaima Zarnaz
There are several circumstances in which constantly updated multidimensional tensor data must be analyzed in real-time in order to yield quick recommendation in our fast-changing data world. Methods for incremental tensor decompositions are powerful tools for analyzing and predicting fast-growing multidimensional datasets. In this research, we provide a robust model that overcomes the limitations of the most popular incremental tensor decomposition methods and yields high-accuracy prediction results for enormous datasets with fast execution time. Testing our model on datasets, we discovered that it was able to create a tensor summary that could reflect both the new and old datasets properly and performed better than traditional static methods.
{"title":"Progressive Recommendation by Incremental Tensor Factorization","authors":"Dipannita Biswas, K. M. Azharul Hasan, Zaima Zarnaz","doi":"10.1109/ICCIT57492.2022.10054697","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10054697","url":null,"abstract":"There are several circumstances in which constantly updated multidimensional tensor data must be analyzed in real-time in order to yield quick recommendation in our fast-changing data world. Methods for incremental tensor decompositions are powerful tools for analyzing and predicting fast-growing multidimensional datasets. In this research, we provide a robust model that overcomes the limitations of the most popular incremental tensor decomposition methods and yields high-accuracy prediction results for enormous datasets with fast execution time. Testing our model on datasets, we discovered that it was able to create a tensor summary that could reflect both the new and old datasets properly and performed better than traditional static methods.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127395464","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 : 2022-12-17DOI: 10.1109/ICCIT57492.2022.10055575
M. Hasan, Prince Mahmud, Mst. Merina Khatun, Md. Hasibur Rahman, Md. Tarequl Islam
With the exponential growth of biological databases, it is a highly challenging and essential task to find the exact string from these databases. Various string pattern searching strategies played an essential role in solving string search problems. Mainly, the string pattern matching algorithm aims to reduce the running time which depends on attempts and character comparisons. Our paper proposed a new hybrid strategy, the Largest Shift Algorithm (LSA), to solve these string pattern matching problems more effectively. Our suggested new hybrid algorithm combines the most advantageous characteristics of Berry Ravindran, Zhu-Takaoka, and a customized Back & Forth Matching (BFM) algorithm. These three algorithms were chosen as they perform better in the tests for counting attempts and character comparisons. Three distinct types of algorithms were tested to analyze the performance of our proposed LSA algorithm, discussed in the literature which are BRR, maximum-shift, and quick-search algorithms. We have used English text, DNA, and protein sequences as data for their different nature. The proposed algorithm outperforms the previous algorithms in terms of performance such as runtime, total attempts, and comparisons of character.
{"title":"Largest Shift String Matching Algorithm: Blend of Berry Ravindran, Zhu-Takaoka and Back & Forth Matching Algorithm","authors":"M. Hasan, Prince Mahmud, Mst. Merina Khatun, Md. Hasibur Rahman, Md. Tarequl Islam","doi":"10.1109/ICCIT57492.2022.10055575","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10055575","url":null,"abstract":"With the exponential growth of biological databases, it is a highly challenging and essential task to find the exact string from these databases. Various string pattern searching strategies played an essential role in solving string search problems. Mainly, the string pattern matching algorithm aims to reduce the running time which depends on attempts and character comparisons. Our paper proposed a new hybrid strategy, the Largest Shift Algorithm (LSA), to solve these string pattern matching problems more effectively. Our suggested new hybrid algorithm combines the most advantageous characteristics of Berry Ravindran, Zhu-Takaoka, and a customized Back & Forth Matching (BFM) algorithm. These three algorithms were chosen as they perform better in the tests for counting attempts and character comparisons. Three distinct types of algorithms were tested to analyze the performance of our proposed LSA algorithm, discussed in the literature which are BRR, maximum-shift, and quick-search algorithms. We have used English text, DNA, and protein sequences as data for their different nature. The proposed algorithm outperforms the previous algorithms in terms of performance such as runtime, total attempts, and comparisons of character.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115728271","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 : 2022-12-17DOI: 10.1109/ICCIT57492.2022.10054896
Azizul Haque, Tanzim Hossain, M. N. Murshed, K. I. B. Iqbal, Mohammad Monir Uddin
Supervised learning extracts a relationship between the input and the output from a training dataset. We consider four models – Support Vector Machine, Random Forest, Gradient Boost, and K-Nearest Neighbor – and employ them on data pertaining to airfoils in two different cases. First, given data about several different airfoil configurations, our objective is to predict the aerodynamic coefficients of a new airfoil at different angles of attack. Second, we seek to investigate how the coefficients can be estimated for a specific airfoil if the Reynolds number dramatically changes. It is our finding that the Random Forest and the Gradient Boost show promising performance in both the scenarios.
{"title":"Estimating Aerodynamic Data via Supervised Learning","authors":"Azizul Haque, Tanzim Hossain, M. N. Murshed, K. I. B. Iqbal, Mohammad Monir Uddin","doi":"10.1109/ICCIT57492.2022.10054896","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10054896","url":null,"abstract":"Supervised learning extracts a relationship between the input and the output from a training dataset. We consider four models – Support Vector Machine, Random Forest, Gradient Boost, and K-Nearest Neighbor – and employ them on data pertaining to airfoils in two different cases. First, given data about several different airfoil configurations, our objective is to predict the aerodynamic coefficients of a new airfoil at different angles of attack. Second, we seek to investigate how the coefficients can be estimated for a specific airfoil if the Reynolds number dramatically changes. It is our finding that the Random Forest and the Gradient Boost show promising performance in both the scenarios.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115782403","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：