Pub Date : 2018-12-01DOI: 10.1109/ICCITECHN.2018.8631973
Sadia Jahan, Swakkhar Shatabda, D. Farid
Active learning also known as an optimal experimental design, is a process for building a classifier or learning model with less number of training instances in the semi-supervised setting. It's a well-known approach that is used in many real-life machine learning and data mining applications. Active learning uses a query function and an oracle or expert (e.g., a human or information source) for labeling unlabeled data instances to boost up the performance of a classifier. Labeling the unlabeled data instances is difficult, time-consuming, and expensive. In this paper, we have proposed an approach based on cluster analysis for selecting informative training instances from large number of unlabeled data instances or big data that helps us to select less number of training instances to build a classifier suitable for active learning. The proposed method clusters the unlabeled big data into several clusters and find the informative instances from each cluster based on the center of the cluster, nearest neighbors of the center of the cluster, and also selecting random instances from each cluster. The objective is to find the informative unlabeled instances and label them by the oracle for scaling up the classification results of the machine learning algorithms to be applied on big data. We have tested the performance of the proposed method on seven benchmark datasets from UC Irvine Machine Learning Repository employing following five well-known machine learning algorithms: C4.5 (decision tree induction), SVM (support vector machines), Random Forest, Bagging, and Boosting (AdaBoost). The experimental analysis proved that proposed method improves the performance of classifiers in active learning with less number of training instances.
{"title":"Active Learning for Mining Big Data","authors":"Sadia Jahan, Swakkhar Shatabda, D. Farid","doi":"10.1109/ICCITECHN.2018.8631973","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631973","url":null,"abstract":"Active learning also known as an optimal experimental design, is a process for building a classifier or learning model with less number of training instances in the semi-supervised setting. It's a well-known approach that is used in many real-life machine learning and data mining applications. Active learning uses a query function and an oracle or expert (e.g., a human or information source) for labeling unlabeled data instances to boost up the performance of a classifier. Labeling the unlabeled data instances is difficult, time-consuming, and expensive. In this paper, we have proposed an approach based on cluster analysis for selecting informative training instances from large number of unlabeled data instances or big data that helps us to select less number of training instances to build a classifier suitable for active learning. The proposed method clusters the unlabeled big data into several clusters and find the informative instances from each cluster based on the center of the cluster, nearest neighbors of the center of the cluster, and also selecting random instances from each cluster. The objective is to find the informative unlabeled instances and label them by the oracle for scaling up the classification results of the machine learning algorithms to be applied on big data. We have tested the performance of the proposed method on seven benchmark datasets from UC Irvine Machine Learning Repository employing following five well-known machine learning algorithms: C4.5 (decision tree induction), SVM (support vector machines), Random Forest, Bagging, and Boosting (AdaBoost). The experimental analysis proved that proposed method improves the performance of classifiers in active learning with less number of training instances.","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134343952","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 : 2018-12-01DOI: 10.1109/ICCITECHN.2018.8631934
Avik Sarkar, M. Hossen
With the increasing amount of data within the cloud, it is harder to get the expected one. This leads to the idea of text summarization. Automatic text summarization is a tool for summarizing textual data into a short and concise piece of information via which people can have the idea about the content. Several approaches are introduced but there are a little amount of work has been done on Bangla text summarizing techniques due to some different and multifaceted structure of Bangla language. This paper illustrates the implementation of term frequency and semantic sentence similarity based summarizing approaches to summarize a single Bangla document. Removing stopwords, noisy words, lemmatization, tokenization has been done beforehand. Both of these methods return a bunch of top-ranked sentences to create a summary. The rank of a sentence is determined by the term frequency for the first approach and the sentence similarity for the second approach. The experimental result shows a favorable outcome for both of the approaches. Further improvements of these approaches certainly will return an enchanting outcome.
{"title":"Automatic Bangla Text Summarization Using Term Frequency and Semantic Similarity Approach","authors":"Avik Sarkar, M. Hossen","doi":"10.1109/ICCITECHN.2018.8631934","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631934","url":null,"abstract":"With the increasing amount of data within the cloud, it is harder to get the expected one. This leads to the idea of text summarization. Automatic text summarization is a tool for summarizing textual data into a short and concise piece of information via which people can have the idea about the content. Several approaches are introduced but there are a little amount of work has been done on Bangla text summarizing techniques due to some different and multifaceted structure of Bangla language. This paper illustrates the implementation of term frequency and semantic sentence similarity based summarizing approaches to summarize a single Bangla document. Removing stopwords, noisy words, lemmatization, tokenization has been done beforehand. Both of these methods return a bunch of top-ranked sentences to create a summary. The rank of a sentence is determined by the term frequency for the first approach and the sentence similarity for the second approach. The experimental result shows a favorable outcome for both of the approaches. Further improvements of these approaches certainly will return an enchanting outcome.","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133576921","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 : 2018-12-01DOI: 10.1109/ICCITECHN.2018.8631930
M. Raihan, M. M. Islam, Promila Ghosh, S. Shaj, Mubtasim Rafid Chowdhury, Saikat Mondal, A. More
Acute Coronary Syndrome (ACS) is liable for the sudden death. The originator of tachycardia is drug addiction, hyperpiesia polygenic disorder, lipidemia. From the healthcare unit, ACS patients dataset has been collected. By preprocessing the information the chances of the exigency of tachycardia by possessing machine learning (ML) approaches are analyzed. The proficiency of ML techniques for prediction is authentic than any other traditional systems. The central scheme of this analysis is to anticipate the significant contingency of tachycardia. Neural Network, SVM, AdaBoost, Bagging, K-NN, Random Forest approaches are used as long as anticipating the betrayal of ACS. The high-grade exactness with AdaBoost and Bagging are 75.49% and 76.28%. The precision and recall for AdaBoost are 0.741; 0.75 and 0.755; 0.763 for Bagging techniques respectively.
{"title":"A Comprehensive Analysis on Risk Prediction of Acute Coronary Syndrome Using Machine Learning Approaches","authors":"M. Raihan, M. M. Islam, Promila Ghosh, S. Shaj, Mubtasim Rafid Chowdhury, Saikat Mondal, A. More","doi":"10.1109/ICCITECHN.2018.8631930","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631930","url":null,"abstract":"Acute Coronary Syndrome (ACS) is liable for the sudden death. The originator of tachycardia is drug addiction, hyperpiesia polygenic disorder, lipidemia. From the healthcare unit, ACS patients dataset has been collected. By preprocessing the information the chances of the exigency of tachycardia by possessing machine learning (ML) approaches are analyzed. The proficiency of ML techniques for prediction is authentic than any other traditional systems. The central scheme of this analysis is to anticipate the significant contingency of tachycardia. Neural Network, SVM, AdaBoost, Bagging, K-NN, Random Forest approaches are used as long as anticipating the betrayal of ACS. The high-grade exactness with AdaBoost and Bagging are 75.49% and 76.28%. The precision and recall for AdaBoost are 0.741; 0.75 and 0.755; 0.763 for Bagging techniques respectively.","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133328419","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 : 2018-12-01DOI: 10.1109/ICCITECHN.2018.8631924
Tasbiraha Athaya, S. Munira, Afsana Zaman, Syed Akhter Hossain, Col A B M Humayun Kabir
Meeting is an important part of our professional life to share and discuss issues based on multiple confidential documents. There are different types of approaches, methods and techniques that are used in organizations for meeting management to track these meetings and store all the meeting files accordingly. In this research work, a new idea of meeting scheduling system is proposed. The objective of this work is to propose two simple algorithms that will detect the conflict of time among the attendees of the meeting and automatically schedule a meeting by using the necessary information. The algorithm is tested in web-based technology. The system is designed to be a secure system to preserve all the documents related to the meeting. In the proposed system, firstly, a meeting list is maintained in a sorted way which helps to schedule meetings without conflict of time. In case of conflict of time, the system proposes available time automatically. Secondly, the option to upload files of the meetings is provided with the security of the files through encoding and decoding. The system is tested with satisfaction and implemented in the university for the use.
{"title":"A Proposed Algorithm and Architecture for Automated Meeting Scheduling and Document Management","authors":"Tasbiraha Athaya, S. Munira, Afsana Zaman, Syed Akhter Hossain, Col A B M Humayun Kabir","doi":"10.1109/ICCITECHN.2018.8631924","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631924","url":null,"abstract":"Meeting is an important part of our professional life to share and discuss issues based on multiple confidential documents. There are different types of approaches, methods and techniques that are used in organizations for meeting management to track these meetings and store all the meeting files accordingly. In this research work, a new idea of meeting scheduling system is proposed. The objective of this work is to propose two simple algorithms that will detect the conflict of time among the attendees of the meeting and automatically schedule a meeting by using the necessary information. The algorithm is tested in web-based technology. The system is designed to be a secure system to preserve all the documents related to the meeting. In the proposed system, firstly, a meeting list is maintained in a sorted way which helps to schedule meetings without conflict of time. In case of conflict of time, the system proposes available time automatically. Secondly, the option to upload files of the meetings is provided with the security of the files through encoding and decoding. The system is tested with satisfaction and implemented in the university for the use.","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114161173","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 : 2018-12-01DOI: 10.1109/ICCITECHN.2018.8631949
Mohammad Woli Ullah, M. A. Rahman, Faisal Shahriar, Zia Uddin Ahmed, Muhammad Mostafa Amir Faisal, Md. Jashim Uddin, Syed Zahidur Rashid
Submarine cable is most preferable technology to provide electricity and broadband internet into Saint Martin of Bangladesh. The most efficient path, its geographical information and the possible threats should be analyzed before installing cable. A geographical survey on Naaf channel and possible landing station (North Beach, Saint Martin and Dokkin Para, Shah Porir Dip) are shown. The route through the seabed and suitable type of submarine power cable are also proposed in the paper. The conditions of the proposed area are suitable for installing cable and threats are not fatal.
海底电缆是向孟加拉国圣马丁提供电力和宽带互联网的最佳技术。在安装电缆之前，应该分析最有效的路径，其地理信息和可能的威胁。对Naaf海峡和可能的登陆站(北滩，圣马丁和Dokkin Para, Shah Porir Dip)的地理调查显示。文中还提出了海底电缆的穿越路线和合适的海底电力电缆类型。拟议区域的条件适合安装电缆，并且威胁不是致命的。
{"title":"Enlightenment of Saint Martin Island: Underwater Submarine Cable and its Reliability","authors":"Mohammad Woli Ullah, M. A. Rahman, Faisal Shahriar, Zia Uddin Ahmed, Muhammad Mostafa Amir Faisal, Md. Jashim Uddin, Syed Zahidur Rashid","doi":"10.1109/ICCITECHN.2018.8631949","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631949","url":null,"abstract":"Submarine cable is most preferable technology to provide electricity and broadband internet into Saint Martin of Bangladesh. The most efficient path, its geographical information and the possible threats should be analyzed before installing cable. A geographical survey on Naaf channel and possible landing station (North Beach, Saint Martin and Dokkin Para, Shah Porir Dip) are shown. The route through the seabed and suitable type of submarine power cable are also proposed in the paper. The conditions of the proposed area are suitable for installing cable and threats are not fatal.","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127550534","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 : 2018-12-01DOI: 10.1109/ICCITECHN.2018.8631952
Nishako Chakma, M. Farhad, Akib Jayed Islam, Swarup Chakraborty, Md Siddat Bin Nesar, Md. Abdul Muktadir
Microstrip patch antennas which have the multiplicity of features are receiving attention for satellite applications nowadays. A microstrip patch antenna operating at K band (18 GHz to 27 GHz) for satellite applications is designed in this paper. The antenna has been designed with CST Microwave Studio Suite software. The miniaturized antenna shows impressive results which are suitable for satellite communications. To implement the antenna for satellite application, prerequisites such as low return loss, high bandwidth, gain, far-field radiation are accomplished here. All the performance parameters are utilized to operate the antenna in K-band. The simulation data show satisfactory results for the above-mentioned applications.
微带贴片天线具有多种特性，在卫星应用中受到广泛关注。设计了一种工作在K波段(18ghz ~ 27ghz)的卫星微带贴片天线。利用CST Microwave Studio Suite软件对天线进行了设计。小型化的天线显示出令人印象深刻的效果，适用于卫星通信。该天线要实现卫星应用，必须具备低回波损耗、高带宽、增益、远场辐射等先决条件。利用所有的性能参数对k波段天线进行操作。仿真数据显示了上述应用的满意结果。
{"title":"Performance Analysis of a Microstrip Patch Antenna for Satellite Communications at K Band","authors":"Nishako Chakma, M. Farhad, Akib Jayed Islam, Swarup Chakraborty, Md Siddat Bin Nesar, Md. Abdul Muktadir","doi":"10.1109/ICCITECHN.2018.8631952","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631952","url":null,"abstract":"Microstrip patch antennas which have the multiplicity of features are receiving attention for satellite applications nowadays. A microstrip patch antenna operating at K band (18 GHz to 27 GHz) for satellite applications is designed in this paper. The antenna has been designed with CST Microwave Studio Suite software. The miniaturized antenna shows impressive results which are suitable for satellite communications. To implement the antenna for satellite application, prerequisites such as low return loss, high bandwidth, gain, far-field radiation are accomplished here. All the performance parameters are utilized to operate the antenna in K-band. The simulation data show satisfactory results for the above-mentioned applications.","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127594116","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 : 2018-12-01DOI: 10.1109/ICCITECHN.2018.8631917
M. Haque
Keyword extraction refers to the identification of words or short phrases that concisely describe the contents of the document. Rapid Automatic Keyword Extraction (RAKE) is a well-known keyword extraction approach. But we found that RAKE fails to extract the significant Bengali keywords. In this paper, we have proposed an improved version of the pristine RAKE called RAKEB. We have also shown that RAKEB works significantly well for Bengali than the pristine RAKE.
{"title":"Automatic Keyword Extraction from Bengali Text Using Improved RAKE Approach","authors":"M. Haque","doi":"10.1109/ICCITECHN.2018.8631917","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631917","url":null,"abstract":"Keyword extraction refers to the identification of words or short phrases that concisely describe the contents of the document. Rapid Automatic Keyword Extraction (RAKE) is a well-known keyword extraction approach. But we found that RAKE fails to extract the significant Bengali keywords. In this paper, we have proposed an improved version of the pristine RAKE called RAKEB. We have also shown that RAKEB works significantly well for Bengali than the pristine RAKE.","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124438283","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 : 2018-12-01DOI: 10.1109/ICCITECHN.2018.8631954
Md. Abdul Kaium Khan, Towqir Ahmed Shaem, Mahbubur Rahman, Abdullah Zowad Khan, M. Alamgir
In this paper, a portable wireless biometric attendance system for academic purpose has been designed and developed using fingerprint sensor and NodeMCU microcontroller. One of the cardinal features of this system is the portability of this system and also the reduction of time consumption compared to the conventional stationary wired/wireless systems where students have to stand in serials and wait for his/her turn which costs good amount of time during a class period. The smaller size of our system has enabled the students to pass the device from one to another without interrupting the class lecture and saves valuable time of the class lecturer which is generally a fixed period of time. All the data of a particular student are stored in a database when a student's finger is registered for a particular institute and after that whenever the student enrolls, his/her attendance is counted automatically and stored in the destined database wirelessly. The attendance marks based on attendance percentages of the students are automatically calculated and uploaded on the webpage. So whenever a semester finishes the course teacher can get percentage information as well as the corresponding attendance marks of all the students without having a headache of it in the entire semester. Our system also notifies the user if any student is absent for 10 successive classes (as per Govt. rule in Bangladesh, course teacher has to inform authority if any student is absent for 10 days successively).
{"title":"A Portable and Less Time Consuming Wireless Biometric Attendance System for Academic Purpose Using NodeMCU Microcontroller","authors":"Md. Abdul Kaium Khan, Towqir Ahmed Shaem, Mahbubur Rahman, Abdullah Zowad Khan, M. Alamgir","doi":"10.1109/ICCITECHN.2018.8631954","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631954","url":null,"abstract":"In this paper, a portable wireless biometric attendance system for academic purpose has been designed and developed using fingerprint sensor and NodeMCU microcontroller. One of the cardinal features of this system is the portability of this system and also the reduction of time consumption compared to the conventional stationary wired/wireless systems where students have to stand in serials and wait for his/her turn which costs good amount of time during a class period. The smaller size of our system has enabled the students to pass the device from one to another without interrupting the class lecture and saves valuable time of the class lecturer which is generally a fixed period of time. All the data of a particular student are stored in a database when a student's finger is registered for a particular institute and after that whenever the student enrolls, his/her attendance is counted automatically and stored in the destined database wirelessly. The attendance marks based on attendance percentages of the students are automatically calculated and uploaded on the webpage. So whenever a semester finishes the course teacher can get percentage information as well as the corresponding attendance marks of all the students without having a headache of it in the entire semester. Our system also notifies the user if any student is absent for 10 successive classes (as per Govt. rule in Bangladesh, course teacher has to inform authority if any student is absent for 10 days successively).","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116650433","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 : 2018-12-01DOI: 10.1109/ICCITECHN.2018.8631964
A. G. Bari, Alessio Gaspar
In previous studies, we proposed four relaxed selections schemes for test-based Pareto coevolution as implemented by a variant of the Population based Pareto Hill Climber (P-PH C- P) Three of them outperformed the default selection used in P-PHC-P in which a parent is only replaced in the next generation by its child if the latter Pareto-dominates the former. While the results were particularly encouraging, more work is needed to fully understand the reasons behind this improved performance. In this work, we therefore extend previous results by revisiting the relaxed selection methods from the perspective of both the distribution of candidate solutions in different Pareto layers, and the concept of hyper volume commonly used in the evolutionary multi-objectives optimization literature. Extensive experimental analysis shows that relaxed selection (Upward-Horizontal Selection) improves convergence while maintaining diversity in the converging population, better than base selection.
{"title":"Investigating Relaxed Selection in Test-Based Pareto Coevolution","authors":"A. G. Bari, Alessio Gaspar","doi":"10.1109/ICCITECHN.2018.8631964","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631964","url":null,"abstract":"In previous studies, we proposed four relaxed selections schemes for test-based Pareto coevolution as implemented by a variant of the Population based Pareto Hill Climber (P-PH C- P) Three of them outperformed the default selection used in P-PHC-P in which a parent is only replaced in the next generation by its child if the latter Pareto-dominates the former. While the results were particularly encouraging, more work is needed to fully understand the reasons behind this improved performance. In this work, we therefore extend previous results by revisiting the relaxed selection methods from the perspective of both the distribution of candidate solutions in different Pareto layers, and the concept of hyper volume commonly used in the evolutionary multi-objectives optimization literature. Extensive experimental analysis shows that relaxed selection (Upward-Horizontal Selection) improves convergence while maintaining diversity in the converging population, better than base selection.","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126784485","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 : 2018-12-01DOI: 10.1109/ICCITECHN.2018.8631966
Saimoom Safayet Akash, Sandip Kabiraz, SM Ashraful Islam, S. Siddique, M. N. Huda, Ishteaque Alam
This paper has developed and demonstrated a system to build traffic instruction detection and translation tools that can extract and convert Bangla text from natural images containing traffic instruction. In the process of developing the system, we have applied various techniques to extract and convert information from natural images. These techniques involve Image Processing, Machine Learning, Optical Character Recognition and Machine Translation. The proposed system consists of three steps, which are Text extraction from image, Post Processing by Language Model and Machine Translation.
{"title":"A Real Time Approach for Bangla Text Extraction and Translation from Traffic Sign","authors":"Saimoom Safayet Akash, Sandip Kabiraz, SM Ashraful Islam, S. Siddique, M. N. Huda, Ishteaque Alam","doi":"10.1109/ICCITECHN.2018.8631966","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631966","url":null,"abstract":"This paper has developed and demonstrated a system to build traffic instruction detection and translation tools that can extract and convert Bangla text from natural images containing traffic instruction. In the process of developing the system, we have applied various techniques to extract and convert information from natural images. These techniques involve Image Processing, Machine Learning, Optical Character Recognition and Machine Translation. The proposed system consists of three steps, which are Text extraction from image, Post Processing by Language Model and Machine Translation.","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117048612","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：