Recommender Systems is a system that helps users to find good stuff like movies, books etc. The user gives ratings to items and it is a system that predicts these ratings. User rates those items that are of their interest. Two types of techniques are used by the recommender systems for recommendations-content based filtering (CBF) and collaborative filtering (CF). Both techniques have their own pros and cons. The most common problems with CF are a cold star, scalability, and sparsity. Also, collaborative filtering needs a large amount of data. We will propose a solution using Kernel Mapping Recommender (KMR) to resolve the recommendation issues like a cold star, scalability, and sparsity the main goal of proposed work is to find dynamic, efficient and effective recommendation algorithm that can be successfully used to make a recommendation to users'. Several heuristic algorithms have been anticipated that merge different categories of KMR for improving correctness and removal of problems linked with a predictable recommender system. The proposed system is checked on movie datasets that are available online and then standardized with KMR. In conditions of accuracy, precision, recall, F1 measure, and ROC metrics; the results expose that the proposed algorithm is quite precise mainly under cold-start and sparse situations.
{"title":"An Efficient Algorithm for Recommender System Using Kernel Mapping Techniques","authors":"Summia Naz, M. Maqsood, Mehr Yahya Durrani","doi":"10.1145/3316615.3316623","DOIUrl":"https://doi.org/10.1145/3316615.3316623","url":null,"abstract":"Recommender Systems is a system that helps users to find good stuff like movies, books etc. The user gives ratings to items and it is a system that predicts these ratings. User rates those items that are of their interest. Two types of techniques are used by the recommender systems for recommendations-content based filtering (CBF) and collaborative filtering (CF). Both techniques have their own pros and cons. The most common problems with CF are a cold star, scalability, and sparsity. Also, collaborative filtering needs a large amount of data. We will propose a solution using Kernel Mapping Recommender (KMR) to resolve the recommendation issues like a cold star, scalability, and sparsity the main goal of proposed work is to find dynamic, efficient and effective recommendation algorithm that can be successfully used to make a recommendation to users'. Several heuristic algorithms have been anticipated that merge different categories of KMR for improving correctness and removal of problems linked with a predictable recommender system. The proposed system is checked on movie datasets that are available online and then standardized with KMR. In conditions of accuracy, precision, recall, F1 measure, and ROC metrics; the results expose that the proposed algorithm is quite precise mainly under cold-start and sparse situations.","PeriodicalId":268392,"journal":{"name":"Proceedings of the 2019 8th International Conference on Software and Computer Applications","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123083240","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}
There are many blockchain frameworks have been introduced to user, developer, and customer and they have to make a decision which one is most suitable for them in order for developing their applications. This paper describes an empirical research to benchmark the properties of frameworks in Hyperledger family. The purpose is to show the advantage & disadvantage of each framework. We apply the deeply study inside the framework and do the experiment to measure the properties as well as give literature and numerical review of each framework. Our results show the status of tested frameworks and do not demonstrate how to employ the framework as well as apply to the detail application. The main contribution of this paper is a statistical analysis of the effectiveness of Hyperledger frameworks as a tool for developer to develop their applications.
{"title":"Survey of Hyperledger Blockchain Frameworks: Case Study in FPT University's Cryptocurrency Wallets","authors":"Tran Quy Ban, Bui Ngoc Anh, Ngo Tung Son, Tran Van Dinh","doi":"10.1145/3316615.3316671","DOIUrl":"https://doi.org/10.1145/3316615.3316671","url":null,"abstract":"There are many blockchain frameworks have been introduced to user, developer, and customer and they have to make a decision which one is most suitable for them in order for developing their applications. This paper describes an empirical research to benchmark the properties of frameworks in Hyperledger family. The purpose is to show the advantage & disadvantage of each framework. We apply the deeply study inside the framework and do the experiment to measure the properties as well as give literature and numerical review of each framework. Our results show the status of tested frameworks and do not demonstrate how to employ the framework as well as apply to the detail application. The main contribution of this paper is a statistical analysis of the effectiveness of Hyperledger frameworks as a tool for developer to develop their applications.","PeriodicalId":268392,"journal":{"name":"Proceedings of the 2019 8th International Conference on Software and Computer Applications","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114287477","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}
V. Mezhuyev, O. M. Lytvyn, I. Pershyna, O. Nechuiviter, O. Lytvyn, V. Lavrik, Oksana Kovalska, Y. Gunchenko
Any software development company encounters issues that need decision-making. At the same time, the use of reliable and proven methods of decision-making does not become a common practice in software companies worldwide. The issues here are the need for extra staff training, allocation of additional time, and inertia of the software industry. This research studies the problems of adoption of the methods of decision-making in the software development companies of Ukraine and Malaysia. The survey was conducted to evaluate software engineers' attitudes toward the use of the methods of decision-making. Research outcomes can be used to raise the level of adoption of the methods of decision-making in software companies worldwide.
{"title":"Acceptance of the Methods of Decision-making: A Case Study from Software Development Companies in Ukraine and Malaysia","authors":"V. Mezhuyev, O. M. Lytvyn, I. Pershyna, O. Nechuiviter, O. Lytvyn, V. Lavrik, Oksana Kovalska, Y. Gunchenko","doi":"10.1145/3316615.3316677","DOIUrl":"https://doi.org/10.1145/3316615.3316677","url":null,"abstract":"Any software development company encounters issues that need decision-making. At the same time, the use of reliable and proven methods of decision-making does not become a common practice in software companies worldwide. The issues here are the need for extra staff training, allocation of additional time, and inertia of the software industry. This research studies the problems of adoption of the methods of decision-making in the software development companies of Ukraine and Malaysia. The survey was conducted to evaluate software engineers' attitudes toward the use of the methods of decision-making. Research outcomes can be used to raise the level of adoption of the methods of decision-making in software companies worldwide.","PeriodicalId":268392,"journal":{"name":"Proceedings of the 2019 8th International Conference on Software and Computer Applications","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117041290","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}
Hammoudeh S. Alamri, K. Z. Zamli, M. Razak, Ahmad Firdaus
The 0/1 Knapsack problem is one of the most popular real-world optimization problems that arise in searching space and finding the most optimum solution. Theoretically, the optimum solution problem of the 0/1 Knapsack requires suitable technique to explore the search space effectively. Practically, as many metaheuristic algorithms, Whale Optimization Algorithm (WOA) may fail in local optimum solution. This paper proposes Opposition-based Whale Optimization Algorithm (OWOA) to optimize solution problem in 0/1 Knapsack. The OWOA has been tested original WOA by using twenty cases of Knapsack problem and against other metaheuristic algorithms such as (CGMA) and HS-Jaya. The experimental results indicate a significant performance of the optimization solution and stabilization with minimal standard deviation value. This shows that the OWOA improved the original version WOA and has promising result in comparison with other existing algorithms.
{"title":"Solving 0/1 Knapsack Problem using Opposition-based Whale Optimization Algorithm (OWOA)","authors":"Hammoudeh S. Alamri, K. Z. Zamli, M. Razak, Ahmad Firdaus","doi":"10.1145/3316615.3316658","DOIUrl":"https://doi.org/10.1145/3316615.3316658","url":null,"abstract":"The 0/1 Knapsack problem is one of the most popular real-world optimization problems that arise in searching space and finding the most optimum solution. Theoretically, the optimum solution problem of the 0/1 Knapsack requires suitable technique to explore the search space effectively. Practically, as many metaheuristic algorithms, Whale Optimization Algorithm (WOA) may fail in local optimum solution. This paper proposes Opposition-based Whale Optimization Algorithm (OWOA) to optimize solution problem in 0/1 Knapsack. The OWOA has been tested original WOA by using twenty cases of Knapsack problem and against other metaheuristic algorithms such as (CGMA) and HS-Jaya. The experimental results indicate a significant performance of the optimization solution and stabilization with minimal standard deviation value. This shows that the OWOA improved the original version WOA and has promising result in comparison with other existing algorithms.","PeriodicalId":268392,"journal":{"name":"Proceedings of the 2019 8th International Conference on Software and Computer Applications","volume":"63 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129576755","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}
The system was designed and developed for the administration, faculty and staff of Isabela State University Cauayan Campus providing an efficient, organized and well-defined Daily Time Record. Aside from that, the software also gives strong security in recording the attendance of the employees due to the device used. The primary objective of the system is to convert the .csv file from the stand-alone face recognition device into a MariaDB database and arranged it into an organized printable output based on the prescribed DTR format of the Civil Service Commission. The proponents have applied the Agile Model (See Fig.1) and utilizes the different developmental platform for developing the project like the MariaDB Database, PHP (Hypertext pre-processor) and Web browser. Also, the Civil Service Commission Daily Time Record Form for designing the printed DTR. Testing was undergone and some errors shown in the printed DTR encountered by some of the employees. It is found out that errors were come from both to the users and to the face recognition device. The reasons behind problems are, users have multiple ID for both the device and the system and users were missed to register their time in their prescribed remark. To be able to have complete data in the DTR, it is suggested that faculty and staff should have one DTR ID and check the device always if it is under time-in or time-out. Despite the problems arisen, the testing still shows the success of the system because the software provides the converted data into a printable and organized output.
{"title":"Using a Hypertext Pre-processor Code in Converting .csv File into a Civil Service Commission Daily Time Record Format Applying Agile Model","authors":"Joel M. Gumiran, Cherry R. Gumiran","doi":"10.1145/3316615.3316690","DOIUrl":"https://doi.org/10.1145/3316615.3316690","url":null,"abstract":"The system was designed and developed for the administration, faculty and staff of Isabela State University Cauayan Campus providing an efficient, organized and well-defined Daily Time Record. Aside from that, the software also gives strong security in recording the attendance of the employees due to the device used. The primary objective of the system is to convert the .csv file from the stand-alone face recognition device into a MariaDB database and arranged it into an organized printable output based on the prescribed DTR format of the Civil Service Commission. The proponents have applied the Agile Model (See Fig.1) and utilizes the different developmental platform for developing the project like the MariaDB Database, PHP (Hypertext pre-processor) and Web browser. Also, the Civil Service Commission Daily Time Record Form for designing the printed DTR. Testing was undergone and some errors shown in the printed DTR encountered by some of the employees. It is found out that errors were come from both to the users and to the face recognition device. The reasons behind problems are, users have multiple ID for both the device and the system and users were missed to register their time in their prescribed remark. To be able to have complete data in the DTR, it is suggested that faculty and staff should have one DTR ID and check the device always if it is under time-in or time-out. Despite the problems arisen, the testing still shows the success of the system because the software provides the converted data into a printable and organized output.","PeriodicalId":268392,"journal":{"name":"Proceedings of the 2019 8th International Conference on Software and Computer Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129869594","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 Human Resource Record Management System developed not only providing an efficient and systematic tool in managing the record and keeping it securely but delivered an automatic computation of the seminars and training attended and service record of the employee using the NBC guidelines policy. The computed points will be utilized for promotion. Moreover, the system also generates an individual account for the employees for quick access of their record and easily update their Personal Data Sheet through Local Area Network or Wide Area Network. The system was designed and developed using MariaDB Database, PHP (Hypertext Pre-processor), Notepad++ and Web Browser. The developed system was utilized and tested following the test plan of Dilky Firdaus, integrating the Design Science Research Model (See Fig. 1). The said model is significantly applied for the success of the development process wherein it includes problem identification and motivation, definition of objectives as a solution, design and demonstration, evaluation and communication. With that, it is found out that the system was successfully complied all necessary requirements, satisfied the objectives, successfully tested and recommended for fully implementation this coming year.
{"title":"Applying Design Science Research in the Development of Human Resource Record Management System with Predictive Analysis through Pointing System","authors":"Cherry R. Gumiran, Joel M. Gumiran","doi":"10.1145/3316615.3316692","DOIUrl":"https://doi.org/10.1145/3316615.3316692","url":null,"abstract":"In Human Resource Record Management System developed not only providing an efficient and systematic tool in managing the record and keeping it securely but delivered an automatic computation of the seminars and training attended and service record of the employee using the NBC guidelines policy. The computed points will be utilized for promotion. Moreover, the system also generates an individual account for the employees for quick access of their record and easily update their Personal Data Sheet through Local Area Network or Wide Area Network. The system was designed and developed using MariaDB Database, PHP (Hypertext Pre-processor), Notepad++ and Web Browser. The developed system was utilized and tested following the test plan of Dilky Firdaus, integrating the Design Science Research Model (See Fig. 1). The said model is significantly applied for the success of the development process wherein it includes problem identification and motivation, definition of objectives as a solution, design and demonstration, evaluation and communication. With that, it is found out that the system was successfully complied all necessary requirements, satisfied the objectives, successfully tested and recommended for fully implementation this coming year.","PeriodicalId":268392,"journal":{"name":"Proceedings of the 2019 8th International Conference on Software and Computer Applications","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124746612","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}
Medical image processing is the most demanding and emerging field which includes the processing of MRI images. Aneurysm can develop in any blood vessel in the body especially in the brain and abdominal aorta. In regards to brain aneurysm, it has been clarified that the MRI is a more preferable detection method to inspect this brain abnormality. This paper experiments MRI brain aneurysm images using image processing techniques in extracting a clearer image of the brain aneurysm. The method proposed includes preprocessing techniques and post processing techniques. Preprocessing techniques involves noise removal function and image enhancement function. The post processing techniques is segmentation and morphological function. All of these techniques are the basic concepts of image processing. Detection and extraction of brain aneurysm from MRI images of the brain is done by using MATLAB software.
{"title":"Brain Aneurysm Extraction in MRI Images","authors":"Rose Hafsah Abd. Rauf, N. A. Ghafar, N. Khalid","doi":"10.1145/3316615.3316696","DOIUrl":"https://doi.org/10.1145/3316615.3316696","url":null,"abstract":"Medical image processing is the most demanding and emerging field which includes the processing of MRI images. Aneurysm can develop in any blood vessel in the body especially in the brain and abdominal aorta. In regards to brain aneurysm, it has been clarified that the MRI is a more preferable detection method to inspect this brain abnormality. This paper experiments MRI brain aneurysm images using image processing techniques in extracting a clearer image of the brain aneurysm. The method proposed includes preprocessing techniques and post processing techniques. Preprocessing techniques involves noise removal function and image enhancement function. The post processing techniques is segmentation and morphological function. All of these techniques are the basic concepts of image processing. Detection and extraction of brain aneurysm from MRI images of the brain is done by using MATLAB software.","PeriodicalId":268392,"journal":{"name":"Proceedings of the 2019 8th International Conference on Software and Computer Applications","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124033344","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}
Cryptography is one of the best methods to keep the information safe from security attack by unauthorized people. At present, many studies had been conducted to generate a more robust cryptographic algorithm to provide high security for data communication. In this study, the authors proposed an enhanced key security of Playfair cipher algorithm using Playfair cipher 16x16 matrix, XOR, two's complement, and bit swapping. The result of the study shows that the processing time for encryption and decryption processes are linearly proportional. It means the longer characters of the key; the more significant times needed to encrypt and decrypt the key. Also, the proposed method is faster than the existing key security of Playfair cipher algorithm. Experimental results also show that the proposed key security algorithm exhibits strong avalanche effect. It was also found that the key cannot be easily cracked by an attacker performing brute-force attack.
{"title":"An Enhanced Key Security of Playfair Cipher Algorithm","authors":"Richard M. Marzan, Ariel M. Sison","doi":"10.1145/3316615.3316689","DOIUrl":"https://doi.org/10.1145/3316615.3316689","url":null,"abstract":"Cryptography is one of the best methods to keep the information safe from security attack by unauthorized people. At present, many studies had been conducted to generate a more robust cryptographic algorithm to provide high security for data communication. In this study, the authors proposed an enhanced key security of Playfair cipher algorithm using Playfair cipher 16x16 matrix, XOR, two's complement, and bit swapping. The result of the study shows that the processing time for encryption and decryption processes are linearly proportional. It means the longer characters of the key; the more significant times needed to encrypt and decrypt the key. Also, the proposed method is faster than the existing key security of Playfair cipher algorithm. Experimental results also show that the proposed key security algorithm exhibits strong avalanche effect. It was also found that the key cannot be easily cracked by an attacker performing brute-force attack.","PeriodicalId":268392,"journal":{"name":"Proceedings of the 2019 8th International Conference on Software and Computer Applications","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114080802","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}
Modernization and urban development have destroyed the original image of some architectural heritage. One example is the Prophet Mosque in Madinah, where, in its current state, has been through many modifications and expansion from the original form that is built by the prophet. As the original form of the mosque presents a significant historic origin and the culture surrounding that era, this project presents an approach to preserve the original form of this heritage using 3D imagery - particularly virtual reality in disseminating its archaeological, historical facts and heritage awareness for the global community. Firstly, we give an over-view of the data acquisition and the 3D reconstruction process of this heritage. Then we present the process of transferring the model into a virtual reality environment for real time navigation experience using a VR headset. This immersive experience for understanding and adding knowledge on the history of this important heritage is the main the contribution of this research and the model itself in 3D form can later be used by other scholars, researchers and many more for academic and/or public purposes.
{"title":"Modelling of the Prophet Mosque in Virtual Reality","authors":"Mohamad Izani Zainal Abidin, A. Razak","doi":"10.1145/3316615.3320117","DOIUrl":"https://doi.org/10.1145/3316615.3320117","url":null,"abstract":"Modernization and urban development have destroyed the original image of some architectural heritage. One example is the Prophet Mosque in Madinah, where, in its current state, has been through many modifications and expansion from the original form that is built by the prophet. As the original form of the mosque presents a significant historic origin and the culture surrounding that era, this project presents an approach to preserve the original form of this heritage using 3D imagery - particularly virtual reality in disseminating its archaeological, historical facts and heritage awareness for the global community. Firstly, we give an over-view of the data acquisition and the 3D reconstruction process of this heritage. Then we present the process of transferring the model into a virtual reality environment for real time navigation experience using a VR headset. This immersive experience for understanding and adding knowledge on the history of this important heritage is the main the contribution of this research and the model itself in 3D form can later be used by other scholars, researchers and many more for academic and/or public purposes.","PeriodicalId":268392,"journal":{"name":"Proceedings of the 2019 8th International Conference on Software and Computer Applications","volume":"306 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116339143","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}
A. Khanom, Sheikh Mastura Farzana, Tahsinur Rahman, I. Mobin
BookCeption is a web-based book recommendation system that allows readers to browse and buy books across multiple platforms at the most affordable price. It fetches data from E-commerce sites like Barnes and Nobles, Amazon, eBay along with Facebook pages and retail bookstore websites. It is a unique book recommender that uses Machine Learning techniques to recommend books as well as offers from other platforms to the registered users. This paper introduces the architecture of the proposed framework, which integrates the book recommendation system with a platform for buying books. For the book recommender discussed here, four different recommendation techniques were used- SVD, Co-clustering, NMF and Deep Learning to determine the best one for the framework. In terms of computational training time, Co-clustering works best with a time of 5.21 minutes and RMSE value of 0.868. However, on the basis of RMSE value, Deep Learning Embedding Model works best with RMSE value of 0.7599 in 8.06 minutes when the number of epochs and the batch size are 20 and 200 respectively.
{"title":"BookCeption","authors":"A. Khanom, Sheikh Mastura Farzana, Tahsinur Rahman, I. Mobin","doi":"10.1145/3316615.3316721","DOIUrl":"https://doi.org/10.1145/3316615.3316721","url":null,"abstract":"BookCeption is a web-based book recommendation system that allows readers to browse and buy books across multiple platforms at the most affordable price. It fetches data from E-commerce sites like Barnes and Nobles, Amazon, eBay along with Facebook pages and retail bookstore websites. It is a unique book recommender that uses Machine Learning techniques to recommend books as well as offers from other platforms to the registered users. This paper introduces the architecture of the proposed framework, which integrates the book recommendation system with a platform for buying books. For the book recommender discussed here, four different recommendation techniques were used- SVD, Co-clustering, NMF and Deep Learning to determine the best one for the framework. In terms of computational training time, Co-clustering works best with a time of 5.21 minutes and RMSE value of 0.868. However, on the basis of RMSE value, Deep Learning Embedding Model works best with RMSE value of 0.7599 in 8.06 minutes when the number of epochs and the batch size are 20 and 200 respectively.","PeriodicalId":268392,"journal":{"name":"Proceedings of the 2019 8th International Conference on Software and Computer Applications","volume":"47 11","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113985668","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：