Pub Date : 2022-12-17DOI: 10.1109/ICCIT57492.2022.10054976
Abrar Shahriar Abeed, Asif Atiq, Afra Antara Anjum, Azher Ahmed Efat, Dewan Ziaul Karim
Bone Marrow is responsible for the creation of all the different types of blood cells in the human body and helps us to identify various types of bone marrow cell disorders. Therefore it is necessary to correctly identify and classify the different types of cells. Conducting different pathological and blood tests may take some time. Applying a Deep Neural Network (DNN) for blood cell detection allows us to quickly classify the call types, which further enables us to identify multiple types of blood cells simultaneously from the same sample. Not only does this save us the time needed for cell classification but also removes the possibility of human error as an automated system can deliver more precise and instantaneous results than a hematologist or pathologist. Machine Learning algorithms are capable of solving these problems quite easily. With that in mind, we propose a CNN-based architecture named BoMaCNet, which is capable of detecting and classifying bone marrow cell images quickly and accurately. Our CNN model takes 96000 images in total, which are then split into training, testing, and validation. Six common types of bone marrow cells (Artefact, Blast, Erythroblast, Lymphocyte, Segmented Neutrophil and Promyelocyte) are chosen for this research. Our entire data set was split into three parts 80% was kept for training, 10% was kept for validation and 10% was used for testing. For testing, 1600 instances of each label were used. Our model was able to produce the highest by far results on the used dataset by achieving an overall accuracy of 95.71%. With 95.71% accuracy in training and 93.06% accuracy in validation along with achieving an impressive mean average F-1 score of 0.93, we were able to achieve exceptional results.
{"title":"BoMaCNet: A Convolutional Neural Network Model to Detect Bone Marrow Cell Cytology","authors":"Abrar Shahriar Abeed, Asif Atiq, Afra Antara Anjum, Azher Ahmed Efat, Dewan Ziaul Karim","doi":"10.1109/ICCIT57492.2022.10054976","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10054976","url":null,"abstract":"Bone Marrow is responsible for the creation of all the different types of blood cells in the human body and helps us to identify various types of bone marrow cell disorders. Therefore it is necessary to correctly identify and classify the different types of cells. Conducting different pathological and blood tests may take some time. Applying a Deep Neural Network (DNN) for blood cell detection allows us to quickly classify the call types, which further enables us to identify multiple types of blood cells simultaneously from the same sample. Not only does this save us the time needed for cell classification but also removes the possibility of human error as an automated system can deliver more precise and instantaneous results than a hematologist or pathologist. Machine Learning algorithms are capable of solving these problems quite easily. With that in mind, we propose a CNN-based architecture named BoMaCNet, which is capable of detecting and classifying bone marrow cell images quickly and accurately. Our CNN model takes 96000 images in total, which are then split into training, testing, and validation. Six common types of bone marrow cells (Artefact, Blast, Erythroblast, Lymphocyte, Segmented Neutrophil and Promyelocyte) are chosen for this research. Our entire data set was split into three parts 80% was kept for training, 10% was kept for validation and 10% was used for testing. For testing, 1600 instances of each label were used. Our model was able to produce the highest by far results on the used dataset by achieving an overall accuracy of 95.71%. With 95.71% accuracy in training and 93.06% accuracy in validation along with achieving an impressive mean average F-1 score of 0.93, we were able to achieve exceptional results.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"62 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":"115941786","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/ICCIT57492.2022.10054899
Md. Abdul Al Emon, Faria Alam, Ramiza Rumaisa Aliya, Tanha Tasfia, Muhaimin Bin Munir, Ishraq Hasan
A surgeon needs the assistance of a nurse to provide him with the required surgical equipment while conducting surgery. However, in some situations, such as in the case of rural areas or emergencies, there may be a lack of skilled nurses or the nurses may be exhausted due to longer working hours, resulting which there may be possibility of errors at that crucial moment when the accuracy is required the most. Therefore, it may be helpful for the surgeons if a faster and more accurate assistive technology may be introduced, which will be reliable in any situation, and thus would reduce the chances of any mistakes. This work aims at introducing an intelligent robotic arm that tries to meet the requirement of such assistance. The robotic arm introduced here acts according to the voice commands provided by the surgeon. Whenever the arm gets provided with the name of surgical equipment, it will be able to find and hand it over to the doctor, and it will also be able to keep the equipment back in place if commanded so. The proposed solution provides a better way to deal with the problem, as it can work for hours continuously with less chance of errors during surgery. Again, in case of getting handled by nurses, there is also a chance of contamination of the surgical equipment, which our proposed work reduces significantly.
{"title":"Voice-Controlled Intelligent Robotic Arm to Assist Surgeon in Performing Surgery","authors":"Md. Abdul Al Emon, Faria Alam, Ramiza Rumaisa Aliya, Tanha Tasfia, Muhaimin Bin Munir, Ishraq Hasan","doi":"10.1109/ICCIT57492.2022.10054899","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10054899","url":null,"abstract":"A surgeon needs the assistance of a nurse to provide him with the required surgical equipment while conducting surgery. However, in some situations, such as in the case of rural areas or emergencies, there may be a lack of skilled nurses or the nurses may be exhausted due to longer working hours, resulting which there may be possibility of errors at that crucial moment when the accuracy is required the most. Therefore, it may be helpful for the surgeons if a faster and more accurate assistive technology may be introduced, which will be reliable in any situation, and thus would reduce the chances of any mistakes. This work aims at introducing an intelligent robotic arm that tries to meet the requirement of such assistance. The robotic arm introduced here acts according to the voice commands provided by the surgeon. Whenever the arm gets provided with the name of surgical equipment, it will be able to find and hand it over to the doctor, and it will also be able to keep the equipment back in place if commanded so. The proposed solution provides a better way to deal with the problem, as it can work for hours continuously with less chance of errors during surgery. Again, in case of getting handled by nurses, there is also a chance of contamination of the surgical equipment, which our proposed work reduces significantly.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"18 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":"116788771","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/ICCIT57492.2022.10055758
Refaat Mohammad Alamgir, Ali Abir Shuvro, Mueeze Al Mushabbir, Mohammed Ashfaq Raiyan, Nusrat Jahan Rani, M. Rahman, M. H. Kabir, Sabbir Ahmed
The task of locating and classifying different types of vehicles has become a vital element in numerous applications of automation and intelligent systems ranging from traffic surveillance to vehicle identification and many more. In recent times, Deep Learning models have been dominating the field of vehicle detection. Yet, Bangladeshi vehicle detection has remained a relatively unexplored area. One of the main goals of vehicle detection is its real-time application, where ‘You Only Look Once’ (YOLO) models have proven to be the most effective architecture. In this work, intending to find the best-suited YOLO architecture for fast and accurate vehicle detection from traffic images in Bangladesh, we have conducted a performance analysis of different variants of the YOLO-based architectures such as YOLOV3, YOLOV5s, and YOLOV5x. The models were trained on a dataset containing 7390 images belonging to 21 types of vehicles comprising samples from the DhakaAI dataset, the Poribohon-BD dataset, and our self-collected images. After thorough quantitative and qualitative analysis, we found the YOLOV5x variant to be the best-suited model, performing better than YOLOv3 and YOLOv5s models respectively by 7 & 4 percent in mAP, and 12 & 8.5 percent in terms of Accuracy.
{"title":"Performance Analysis of YOLO-based Architectures for Vehicle Detection from Traffic Images in Bangladesh","authors":"Refaat Mohammad Alamgir, Ali Abir Shuvro, Mueeze Al Mushabbir, Mohammed Ashfaq Raiyan, Nusrat Jahan Rani, M. Rahman, M. H. Kabir, Sabbir Ahmed","doi":"10.1109/ICCIT57492.2022.10055758","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10055758","url":null,"abstract":"The task of locating and classifying different types of vehicles has become a vital element in numerous applications of automation and intelligent systems ranging from traffic surveillance to vehicle identification and many more. In recent times, Deep Learning models have been dominating the field of vehicle detection. Yet, Bangladeshi vehicle detection has remained a relatively unexplored area. One of the main goals of vehicle detection is its real-time application, where ‘You Only Look Once’ (YOLO) models have proven to be the most effective architecture. In this work, intending to find the best-suited YOLO architecture for fast and accurate vehicle detection from traffic images in Bangladesh, we have conducted a performance analysis of different variants of the YOLO-based architectures such as YOLOV3, YOLOV5s, and YOLOV5x. The models were trained on a dataset containing 7390 images belonging to 21 types of vehicles comprising samples from the DhakaAI dataset, the Poribohon-BD dataset, and our self-collected images. After thorough quantitative and qualitative analysis, we found the YOLOV5x variant to be the best-suited model, performing better than YOLOv3 and YOLOv5s models respectively by 7 & 4 percent in mAP, and 12 & 8.5 percent in terms of Accuracy.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"29 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":"124899987","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}
In recent times, technological advancement has brought a tremendous change in the field o f c ricket, which is a popular sport in many countries. Technology is being utilized to figure out projected score prediction, wicket prediction, winning probability, run rate, and many other parameters. In this research, our primary goal is to use Machine learning in the field of Cricket, where we aim to classify the shot played by the batsman, which can help in applications such as automated broadcasting systems or statistical data generation systems. For implementing our proposed model, we have generated our own dataset of cricket shot images by taking real-time photos from various cricket matches. We collected 1000 images of 10 different types of shots being played. For the classification task, we trained VGG-19 and Inception v3 model architecture and we got a better result by using VGG19. Before classification, the images had to go through several pre-processing methods such as background removal through Mask R-CNN, batsman segmentation through YOLO v3, etc. Then we used 83% of the total images to train the models and 17% to test the models. Finally, we achieved desired accuracy of 84.71% from VGG-19 and 82.35% from Inception-V3.
{"title":"An Approach to Classify the Shot Selection by Batsmen in Cricket Matches Using Deep Neural Network on Image Data","authors":"Afsana Khan, Fariha Haque Nabila, Masud Mohiuddin, Mahadi Mollah, Ashraful Alam, Md Tanzim Reza","doi":"10.1109/ICCIT57492.2022.10055811","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10055811","url":null,"abstract":"In recent times, technological advancement has brought a tremendous change in the field o f c ricket, which is a popular sport in many countries. Technology is being utilized to figure out projected score prediction, wicket prediction, winning probability, run rate, and many other parameters. In this research, our primary goal is to use Machine learning in the field of Cricket, where we aim to classify the shot played by the batsman, which can help in applications such as automated broadcasting systems or statistical data generation systems. For implementing our proposed model, we have generated our own dataset of cricket shot images by taking real-time photos from various cricket matches. We collected 1000 images of 10 different types of shots being played. For the classification task, we trained VGG-19 and Inception v3 model architecture and we got a better result by using VGG19. Before classification, the images had to go through several pre-processing methods such as background removal through Mask R-CNN, batsman segmentation through YOLO v3, etc. Then we used 83% of the total images to train the models and 17% to test the models. Finally, we achieved desired accuracy of 84.71% from VGG-19 and 82.35% from Inception-V3.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"14 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":"125359419","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/ICCIT57492.2022.10055021
Ashek Seum, Md. Reasad Zaman Chowdhury, F. S. Hossain
As outsourcing of IC manufacturing has become a global phenomenon, the risk of ICs being infested with Trojans has increased more than ever. In this paper, we propose a circuit gate level netlist based Trojan detection using supervised machine learning approaches. A number of features are extracted from the netlist that delivers a sophisticated dataset for the approximately trained model to deliver a higher true positive detection rate. The netlist is analyzed in 45 nm technology to generate features and Monte Carlo simulation is performed to generate two thousand virtual netlists, including one thousand Trojan infested netlists. We experiment with the s27 benchmark circuit netlist to evaluate our approach. Different types of sequential and combinational types of Trojans from literature are inserted into the netlist to evaluate the proposed approach. The results show significant Trojan detectability in different machine learning approaches.
{"title":"An Efficient Machine Learning Approach for Hardware Trojan Detection","authors":"Ashek Seum, Md. Reasad Zaman Chowdhury, F. S. Hossain","doi":"10.1109/ICCIT57492.2022.10055021","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10055021","url":null,"abstract":"As outsourcing of IC manufacturing has become a global phenomenon, the risk of ICs being infested with Trojans has increased more than ever. In this paper, we propose a circuit gate level netlist based Trojan detection using supervised machine learning approaches. A number of features are extracted from the netlist that delivers a sophisticated dataset for the approximately trained model to deliver a higher true positive detection rate. The netlist is analyzed in 45 nm technology to generate features and Monte Carlo simulation is performed to generate two thousand virtual netlists, including one thousand Trojan infested netlists. We experiment with the s27 benchmark circuit netlist to evaluate our approach. Different types of sequential and combinational types of Trojans from literature are inserted into the netlist to evaluate the proposed approach. The results show significant Trojan detectability in different machine learning approaches.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"17 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":"127601537","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/ICCIT57492.2022.10054997
M. M. Rakibul Hasan, Rihab Rahman, Kaisary Zaman
Information Security frameworks can be denoted as the heart of protecting the information of any system. These frameworks ensure information security on a very broad scale, thus adopting them should also be sustainable. University automation system is one of such large-scale system having information of different confidentiality. The shared resources of a university automation system have been identified in this research, and as a result, a framework for information security has been proposed to ensure the system's overall safety.
{"title":"Design an Information Security Framework for University Automation System","authors":"M. M. Rakibul Hasan, Rihab Rahman, Kaisary Zaman","doi":"10.1109/ICCIT57492.2022.10054997","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10054997","url":null,"abstract":"Information Security frameworks can be denoted as the heart of protecting the information of any system. These frameworks ensure information security on a very broad scale, thus adopting them should also be sustainable. University automation system is one of such large-scale system having information of different confidentiality. The shared resources of a university automation system have been identified in this research, and as a result, a framework for information security has been proposed to ensure the system's overall safety.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"13 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":"127934539","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/ICCIT57492.2022.10054766
Rahul Deb Mohalder, Ferdous bin Ali, Laboni Paul, Kamrul Hasan Talukder
Colorectal cancer is one of the deadliest diseases and one of the most difficult diseases to diagnose. A big reason for this is that it takes a long time to identify at an early stage. For treatment, a rapid and precise diagnosis of nodules is very crucial. In order to identify cancer in its early stages, a variety of techniques have been employed. Deep learning approaches were used in this work in order to identify Colorectal cancer tumors. In our this research, we used a dataset of same dimension of colon cancer tissues histopathological images. We proposed a deep learning model for predicting CRC tumors from histopathological images. CNN technique used for analyzing complex data. By CNN technique we analyzed our complex tumor images for identifying abnormal or suspicious tumor patterns. We made a five-layer deep neural network model. It consists of the input layer, four hidden layers, and the output layer. We used Rectified linear unit (ReLU) activation function in the hidden layer and the Softmax function in the output layer. We obtained an accuracy 99.70% from our deep learning model and our model loss was 0.0160. We calculate precision, recall, and F-score for the performance evaluation of our method. It is evident from our experiment that our proposed model produces a better result than some related works.
{"title":"Deep Learning-Based Colon Cancer Tumor Prediction Using Histopathological Images","authors":"Rahul Deb Mohalder, Ferdous bin Ali, Laboni Paul, Kamrul Hasan Talukder","doi":"10.1109/ICCIT57492.2022.10054766","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10054766","url":null,"abstract":"Colorectal cancer is one of the deadliest diseases and one of the most difficult diseases to diagnose. A big reason for this is that it takes a long time to identify at an early stage. For treatment, a rapid and precise diagnosis of nodules is very crucial. In order to identify cancer in its early stages, a variety of techniques have been employed. Deep learning approaches were used in this work in order to identify Colorectal cancer tumors. In our this research, we used a dataset of same dimension of colon cancer tissues histopathological images. We proposed a deep learning model for predicting CRC tumors from histopathological images. CNN technique used for analyzing complex data. By CNN technique we analyzed our complex tumor images for identifying abnormal or suspicious tumor patterns. We made a five-layer deep neural network model. It consists of the input layer, four hidden layers, and the output layer. We used Rectified linear unit (ReLU) activation function in the hidden layer and the Softmax function in the output layer. We obtained an accuracy 99.70% from our deep learning model and our model loss was 0.0160. We calculate precision, recall, and F-score for the performance evaluation of our method. It is evident from our experiment that our proposed model produces a better result than some related works.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"1 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":"128555489","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/ICCIT57492.2022.10054884
Khalid Shifullah, H. M. Rakibullah, Nuzhat Islam, Hasin Raihan, Md. Ashik Iqbal, Dewan Ziaul Karim, Annajiat Alim Rasel
Social media has become an essential part for people all over the world. It has given a platform for people to share thoughts, emotions, opinions, and ideas, causing a huge deal of data upsurge. Such an amount of data could be analyzed based on sentiment analysis and text classification via construction of an effective machine learning model. The concept gets more insight into it through analysis of the data, which is nearly impossible to conduct manually due to its huge configuration. This research focuses on the user’s comments, and reviews about different hotels to predict their sentiment. As for the datasets, comments and reviews of hotels from online sites have been utilized. Moreover, text pre-processing techniques like tokenization, case folding, stopword removal, lemmatization, and duplicate data removal have been applied. TF-IDF and Bag of Words have been applied for word embedding. Furthermore, the effectiveness of supervised machine learning algorithms like, Support Vector Machine, Naïve Bayes, Random Forest, and Logistic Regression was evaluated and from the comparative analysis, it was observed that the Logistic Regression provided the most accuracy ranging from 86 to 89 percent.
社交媒体已经成为世界各地人们不可或缺的一部分。它为人们提供了一个分享思想、情感、观点和想法的平台，引起了巨大的数据热潮。通过构建有效的机器学习模型，可以基于情感分析和文本分类对如此大量的数据进行分析。这个概念通过对数据的分析得到了更深入的了解，由于其庞大的配置，这几乎是不可能手动进行的。本研究的重点是用户的评论，以及对不同酒店的评论，以预测他们的情绪。对于数据集，我们利用了在线网站对酒店的评论和评论。此外，还应用了文本预处理技术，如标记化、案例折叠、停止词删除、词序化和重复数据删除。应用TF-IDF和Bag of Words进行词嵌入。此外，评估了监督机器学习算法(如支持向量机，Naïve贝叶斯，随机森林和逻辑回归)的有效性，并从比较分析中观察到逻辑回归提供了最高的准确性，范围从86%到89%。
{"title":"Classification of Hotel Reviews Using Sentiment Analysis and Machine Learning","authors":"Khalid Shifullah, H. M. Rakibullah, Nuzhat Islam, Hasin Raihan, Md. Ashik Iqbal, Dewan Ziaul Karim, Annajiat Alim Rasel","doi":"10.1109/ICCIT57492.2022.10054884","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10054884","url":null,"abstract":"Social media has become an essential part for people all over the world. It has given a platform for people to share thoughts, emotions, opinions, and ideas, causing a huge deal of data upsurge. Such an amount of data could be analyzed based on sentiment analysis and text classification via construction of an effective machine learning model. The concept gets more insight into it through analysis of the data, which is nearly impossible to conduct manually due to its huge configuration. This research focuses on the user’s comments, and reviews about different hotels to predict their sentiment. As for the datasets, comments and reviews of hotels from online sites have been utilized. Moreover, text pre-processing techniques like tokenization, case folding, stopword removal, lemmatization, and duplicate data removal have been applied. TF-IDF and Bag of Words have been applied for word embedding. Furthermore, the effectiveness of supervised machine learning algorithms like, Support Vector Machine, Naïve Bayes, Random Forest, and Logistic Regression was evaluated and from the comparative analysis, it was observed that the Logistic Regression provided the most accuracy ranging from 86 to 89 percent.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"1 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":"129999476","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/ICCIT57492.2022.10056051
Sabbir Ahmed, S. Yesmin, Lata Rani Saha, A. M. Sadat, Mozammel H. A. Khan
The cultivation of crops on land periodically throughout the year is a cropping pattern. This research considered the prediction of major cropping patterns in Bangladesh through only the cultivation-related factors like land, soil, and climate data using Machine Learning techniques. We have considered 52 Upazilas in Bangladesh for data collection which was extracted from the book series Land and Soil Resources Usage Guidelines (in Bangla) published by SRDI, MoA, Dhaka, Bangladesh. The predictor features are a mixture of categorical and numerical data. On the other hand, the number of predicted classes is very large. So, we have used a machine learning model to introduce an effective cropping pattern prediction method that can handle mixed data points with a large number of classes. Machine learning algorithms such as K-nearest neighbors (KNN), Decision Tree (DT), Random Forest Classifier (RFC), XGboost (XGB), and Support Vector Machine (SVM) have been used for cropping pattern prediction. Our models can accurately predict cropping patterns. We have achieved more than 95% accuracy using our dataset for most of the machine learning models that we have used. Also, we have created a back-end and front-end system to use those trained machine learning models easily to predict cropping patterns.
{"title":"Major Cropping Pattern Prediction in Bangladesh from Land, Soil and Climate Data Using Machine Learning Techniques","authors":"Sabbir Ahmed, S. Yesmin, Lata Rani Saha, A. M. Sadat, Mozammel H. A. Khan","doi":"10.1109/ICCIT57492.2022.10056051","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10056051","url":null,"abstract":"The cultivation of crops on land periodically throughout the year is a cropping pattern. This research considered the prediction of major cropping patterns in Bangladesh through only the cultivation-related factors like land, soil, and climate data using Machine Learning techniques. We have considered 52 Upazilas in Bangladesh for data collection which was extracted from the book series Land and Soil Resources Usage Guidelines (in Bangla) published by SRDI, MoA, Dhaka, Bangladesh. The predictor features are a mixture of categorical and numerical data. On the other hand, the number of predicted classes is very large. So, we have used a machine learning model to introduce an effective cropping pattern prediction method that can handle mixed data points with a large number of classes. Machine learning algorithms such as K-nearest neighbors (KNN), Decision Tree (DT), Random Forest Classifier (RFC), XGboost (XGB), and Support Vector Machine (SVM) have been used for cropping pattern prediction. Our models can accurately predict cropping patterns. We have achieved more than 95% accuracy using our dataset for most of the machine learning models that we have used. Also, we have created a back-end and front-end system to use those trained machine learning models easily to predict cropping patterns.","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":"117083583","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/ICCIT57492.2022.10054696
Md. Ariful Islam Arif, Shahriar Kabir, Md Faruk Hussain Khan, Samrat Kumar Dey, Md. Mahbubur Rahman
5G network can provide high speed data transfer with low latency at present days. Network slicing is the prime capability of 5G, where different slices can be utilized for different purposes. Therefore, the network operators can utilize their resources for the users. Machine Learning (ML) or Deep Learning (DL) approach is recently used to address the network issues. Efficient 5G network slicing using ML or DL can provide an effective network. An endeavour has been made to propose an effective 5G network slicing model by applying different ML and DL algorithms. All the methods are adopted in developing the model by data collection, analysis, processing and finally applying the algorithm on the processed dataset. Later the appropriate classifier is determined for the model subjected to accuracy assessment. The dataset collected for use in the research work focuses on type of uses, equipment, technology, day time, duration, guaranteed bit rate (GBR), rate of packet loss, delay budget of packet and slice. The five DL algorithms used are CNN, RNN, LSTM, Bi-LSTM, CNN-LSTM and the four ML algorithms used are XGBoost, RF, NB, SVM. Indeed, among these algorithms, the RNN algorithm has been able to achieve maximum accuracy. The outcome of the research revealed that the suggested model could have an impact on the allocation of precise 5G network slicing.
{"title":"Machine Learning and Deep Learning Based Network Slicing Models for 5G Network","authors":"Md. Ariful Islam Arif, Shahriar Kabir, Md Faruk Hussain Khan, Samrat Kumar Dey, Md. Mahbubur Rahman","doi":"10.1109/ICCIT57492.2022.10054696","DOIUrl":"https://doi.org/10.1109/ICCIT57492.2022.10054696","url":null,"abstract":"5G network can provide high speed data transfer with low latency at present days. Network slicing is the prime capability of 5G, where different slices can be utilized for different purposes. Therefore, the network operators can utilize their resources for the users. Machine Learning (ML) or Deep Learning (DL) approach is recently used to address the network issues. Efficient 5G network slicing using ML or DL can provide an effective network. An endeavour has been made to propose an effective 5G network slicing model by applying different ML and DL algorithms. All the methods are adopted in developing the model by data collection, analysis, processing and finally applying the algorithm on the processed dataset. Later the appropriate classifier is determined for the model subjected to accuracy assessment. The dataset collected for use in the research work focuses on type of uses, equipment, technology, day time, duration, guaranteed bit rate (GBR), rate of packet loss, delay budget of packet and slice. The five DL algorithms used are CNN, RNN, LSTM, Bi-LSTM, CNN-LSTM and the four ML algorithms used are XGBoost, RF, NB, SVM. Indeed, among these algorithms, the RNN algorithm has been able to achieve maximum accuracy. The outcome of the research revealed that the suggested model could have an impact on the allocation of precise 5G network slicing.","PeriodicalId":255498,"journal":{"name":"2022 25th International Conference on Computer and Information Technology (ICCIT)","volume":"103 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":"121136964","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：