Pub Date : 2020-06-01DOI: 10.6025/jdim/2020/18/3/99-108
Rakesh Kumar, Aditi Sharan
E-commerce websites have become main market players in the 21st century due to advancement in the internet technology. Apart from buying products online, customers are also providing reviews on the products purchased by them. These reviews help new customers to buy various products according to their needs, liking, and preferences. However, millions of reviews are added by the customer on a daily basis. To extract meaningful information manually from these huge amounts of reviews is a tough task. So, it is required to develop an automatic analytics tool for the review sentences. Aspect extraction is one of the vital tasks in the process of meaningful information extraction from the products having various entities. In this work, a novel product aspect extraction approach has been proposed which utilize a graphbased technique with the integration of statistical and semantic information. The analysis of experimental results shows that the proposed approach is efficient and effective in comparison to the state of art methods. Subject Categories and Descriptors [H.2.8 Database Applications]: Data mining; [K.4.4 Electronic Commerce] General Terms: E-Commerce, Customer Reviews, Opinion mining
{"title":"A Graph-Based Approach for Aspect Extraction from Online Customer Reviews","authors":"Rakesh Kumar, Aditi Sharan","doi":"10.6025/jdim/2020/18/3/99-108","DOIUrl":"https://doi.org/10.6025/jdim/2020/18/3/99-108","url":null,"abstract":"E-commerce websites have become main market players in the 21st century due to advancement in the internet technology. Apart from buying products online, customers are also providing reviews on the products purchased by them. These reviews help new customers to buy various products according to their needs, liking, and preferences. However, millions of reviews are added by the customer on a daily basis. To extract meaningful information manually from these huge amounts of reviews is a tough task. So, it is required to develop an automatic analytics tool for the review sentences. Aspect extraction is one of the vital tasks in the process of meaningful information extraction from the products having various entities. In this work, a novel product aspect extraction approach has been proposed which utilize a graphbased technique with the integration of statistical and semantic information. The analysis of experimental results shows that the proposed approach is efficient and effective in comparison to the state of art methods. Subject Categories and Descriptors [H.2.8 Database Applications]: Data mining; [K.4.4 Electronic Commerce] General Terms: E-Commerce, Customer Reviews, Opinion mining","PeriodicalId":303976,"journal":{"name":"J. Digit. Inf. Manag.","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128547259","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 : 2020-06-01DOI: 10.6025/jdim/2020/18/3/85-98
Djamila Bouhalouan, Bakhta Nachet, A. Adla
To ensure continuous production in industrial plants, the high valued manufacturing eqipments should be kept in good working conditions. This brings plants to search for means to control and reduce equipment failures. When faults emerge in plants, appropriate actions for fault diagnosis and reparation must be executed promptly and effectively to prevent large costs due to breakdowns. To provide reliable and effective maintenance support, the aid of advanced decision support technology utilizing previous repair experience is of crucial importance for the expert operators as it provides them valuable troubleshooting clues for new faults. Artificial intelligence (AI) technology, particularly, knowledge-based approach is promising for this domain. It captures efficiency of problem solving expertise from the domain experts; guides the expert operators in rapid fault detection and maintenance. This paper focuses on the design and development of a Knowledge-Intensive Decision Support System (KI-DSS) for Manufacturing Equipment Maintenance in industrial plants to support better maintenance decision and improve maintenance efficiency. With integration of casebased Reasoning and ontology, the KiDSS not only carries out data matching retrieval, but also performs semantic associated data access which is important for intelligent knowledge retrieval in decision support system. A case is executed to illustrate the use of the proposed KI-DSS to show the feasibility of our ap proach and the benefit of the ontology support.
{"title":"Knowledge-Intensive Decision Support System for Manufacturing Equipment Maintenance","authors":"Djamila Bouhalouan, Bakhta Nachet, A. Adla","doi":"10.6025/jdim/2020/18/3/85-98","DOIUrl":"https://doi.org/10.6025/jdim/2020/18/3/85-98","url":null,"abstract":"To ensure continuous production in industrial plants, the high valued manufacturing eqipments should be kept in good working conditions. This brings plants to search for means to control and reduce equipment failures. When faults emerge in plants, appropriate actions for fault diagnosis and reparation must be executed promptly and effectively to prevent large costs due to breakdowns. To provide reliable and effective maintenance support, the aid of advanced decision support technology utilizing previous repair experience is of crucial importance for the expert operators as it provides them valuable troubleshooting clues for new faults. Artificial intelligence (AI) technology, particularly, knowledge-based approach is promising for this domain. It captures efficiency of problem solving expertise from the domain experts; guides the expert operators in rapid fault detection and maintenance. This paper focuses on the design and development of a Knowledge-Intensive Decision Support System (KI-DSS) for Manufacturing Equipment Maintenance in industrial plants to support better maintenance decision and improve maintenance efficiency. With integration of casebased Reasoning and ontology, the KiDSS not only carries out data matching retrieval, but also performs semantic associated data access which is important for intelligent knowledge retrieval in decision support system. A case is executed to illustrate the use of the proposed KI-DSS to show the feasibility of our ap proach and the benefit of the ontology support.","PeriodicalId":303976,"journal":{"name":"J. Digit. Inf. Manag.","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131394938","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 : 2020-06-01DOI: 10.6025/jdim/2020/18/3/109-117
M. Ranjbarfard, Shahideh Ahmadi
There are many studies that have applied data mining to banking. However, the lack of proper data mounts a serious obstacle to the employment of data mining techniques by banks. This paper examines previous data mining research in the field of banking to extract all served entities and attributes required for analytical purposes, categorize these attributes and ultimately present a data model for analysis. After analyzing a wide range of data mining applications in banking, 28 entities with 423 attributes were identified and the final proposed entity-relationship model was drawn. Also, a checklist was provided based on the model for auditing data gap in banks and applied to a real case. The results of this paper can be seen as a supportive tool for improving bank‘s business intelligence maturity from the data perspective and enabling managers for analyzing data requirement of information systems. Subject Categories and Descriptors [H.2.8 Database Applications]; Data mining: [D.3.3 Language Constructs and Features]; Data types and structures General Terms: Data Mining, Banking Data, Data Analysis
{"title":"A Study of Data Requirements for Data Mining Applications in Banking","authors":"M. Ranjbarfard, Shahideh Ahmadi","doi":"10.6025/jdim/2020/18/3/109-117","DOIUrl":"https://doi.org/10.6025/jdim/2020/18/3/109-117","url":null,"abstract":"There are many studies that have applied data mining to banking. However, the lack of proper data mounts a serious obstacle to the employment of data mining techniques by banks. This paper examines previous data mining research in the field of banking to extract all served entities and attributes required for analytical purposes, categorize these attributes and ultimately present a data model for analysis. After analyzing a wide range of data mining applications in banking, 28 entities with 423 attributes were identified and the final proposed entity-relationship model was drawn. Also, a checklist was provided based on the model for auditing data gap in banks and applied to a real case. The results of this paper can be seen as a supportive tool for improving bank‘s business intelligence maturity from the data perspective and enabling managers for analyzing data requirement of information systems. Subject Categories and Descriptors [H.2.8 Database Applications]; Data mining: [D.3.3 Language Constructs and Features]; Data types and structures General Terms: Data Mining, Banking Data, Data Analysis","PeriodicalId":303976,"journal":{"name":"J. Digit. Inf. Manag.","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130652737","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 : 2020-04-01DOI: 10.6025/jdim/2020/18/2/57-64
Djasen Tjendry, Wirawan Istiono
{"title":"Is the Binary Search Faster when Two Variables are Added in the Middle of the Data?","authors":"Djasen Tjendry, Wirawan Istiono","doi":"10.6025/jdim/2020/18/2/57-64","DOIUrl":"https://doi.org/10.6025/jdim/2020/18/2/57-64","url":null,"abstract":"","PeriodicalId":303976,"journal":{"name":"J. Digit. Inf. Manag.","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125669336","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 : 2020-04-01DOI: 10.6025/jdim/2020/18/2/65-77
A. Adla
In decision making, the decision-makers frequently employ and perform routine tasks. These processes normally are time-intensive, complex, and in most cases occur regularly. To address this challenge decision makers reuse the already successful decisions. During difficult times, such actions may lead to save time, energy and man-hours, and also result in effective decision making. Memory building depends on how we successfully store earlier knowledge. We through this work introduce a recommender system which is names as BLKBRS which utilized the earlier successful models. In this work we use a case of bank loan and experimented using a semi-structured multiple attribute recommendation environment, and equate the RL-KBRS with a conventional case based reasoning system. RL-KBRS will compensate for lack of experience of young bank consultants, which permits the spread of knowledge distribution to other banks. Subject Categories and Descriptors [H.3] Information Storage and Retrieval; [I.2] Artificial Intelligence General Terms: Memory-based Approach, Information Search, and retrieval, Recommending systems, Case-Based Reasoning
{"title":"Real Estate Loan Knowledge-Based Recommender System","authors":"A. Adla","doi":"10.6025/jdim/2020/18/2/65-77","DOIUrl":"https://doi.org/10.6025/jdim/2020/18/2/65-77","url":null,"abstract":"In decision making, the decision-makers frequently employ and perform routine tasks. These processes normally are time-intensive, complex, and in most cases occur regularly. To address this challenge decision makers reuse the already successful decisions. During difficult times, such actions may lead to save time, energy and man-hours, and also result in effective decision making. Memory building depends on how we successfully store earlier knowledge. We through this work introduce a recommender system which is names as BLKBRS which utilized the earlier successful models. In this work we use a case of bank loan and experimented using a semi-structured multiple attribute recommendation environment, and equate the RL-KBRS with a conventional case based reasoning system. RL-KBRS will compensate for lack of experience of young bank consultants, which permits the spread of knowledge distribution to other banks. Subject Categories and Descriptors [H.3] Information Storage and Retrieval; [I.2] Artificial Intelligence General Terms: Memory-based Approach, Information Search, and retrieval, Recommending systems, Case-Based Reasoning","PeriodicalId":303976,"journal":{"name":"J. Digit. Inf. Manag.","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117103718","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 : 2020-04-01DOI: 10.6025/jdim/2020/18/2/49-56
A. Rahmatulloh, Neng Ika Kurniati, I. Darmawan, Adi Zaenal Asyikin, Deden Witarsyah
Current technological developments change physical paper patterns into digital, which has a very high impact. Positive impact because paper waste is reduced, on the other hand, the rampant copying of digital data raises the amount of plagiarism that is increasing. At present, there are many efforts made by experts to overcome the problem of plagiarism, one of which is by utilizing the winnowing algorithm as a tool to detect plagiarism data. In its development, many optimizing winnowing algorithms used stemming techniques. The most widely used stemmer algorithms include stemmer porter and nazief-adriani. However, there has not been a discussion on the comparison of the effect of performance using stemmer on the winnowing algorithm in measuring the value of plagiarism. So it is necessary to do research on the effect of stemmer algorithms on winnowing algorithms so that the results of plagiarism detection are more optimal. The results of this study indicate that the effect of nazief-adriani stemmer on the winnowing algorithm is superior to the stemmer porter, only decreasing the detection performance of the 0.28% similarity value while the porter stemmer is superior in increasing the processing time to 69% faster. Subject Categories and Descriptors [I.1.2 Algorithms]; [H.3.3 Information Search and Retrieval] General Terms: Plagiarism Detection, Winnowing algorithms, Stemmers
{"title":"Comparison of the Effects Stemmer Porter and Nazief-Adriani on the Performance of Winnowing Algorithms for Measuring Plagiarism","authors":"A. Rahmatulloh, Neng Ika Kurniati, I. Darmawan, Adi Zaenal Asyikin, Deden Witarsyah","doi":"10.6025/jdim/2020/18/2/49-56","DOIUrl":"https://doi.org/10.6025/jdim/2020/18/2/49-56","url":null,"abstract":"Current technological developments change physical paper patterns into digital, which has a very high impact. Positive impact because paper waste is reduced, on the other hand, the rampant copying of digital data raises the amount of plagiarism that is increasing. At present, there are many efforts made by experts to overcome the problem of plagiarism, one of which is by utilizing the winnowing algorithm as a tool to detect plagiarism data. In its development, many optimizing winnowing algorithms used stemming techniques. The most widely used stemmer algorithms include stemmer porter and nazief-adriani. However, there has not been a discussion on the comparison of the effect of performance using stemmer on the winnowing algorithm in measuring the value of plagiarism. So it is necessary to do research on the effect of stemmer algorithms on winnowing algorithms so that the results of plagiarism detection are more optimal. The results of this study indicate that the effect of nazief-adriani stemmer on the winnowing algorithm is superior to the stemmer porter, only decreasing the detection performance of the 0.28% similarity value while the porter stemmer is superior in increasing the processing time to 69% faster. Subject Categories and Descriptors [I.1.2 Algorithms]; [H.3.3 Information Search and Retrieval] General Terms: Plagiarism Detection, Winnowing algorithms, Stemmers","PeriodicalId":303976,"journal":{"name":"J. Digit. Inf. Manag.","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121305333","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 : 2019-12-01DOI: 10.6025/jdim/2019/17/6/337-345
Bilal Benmessahel, Farid Nouioua, Mohamed Touahria, A. Chariete
In this workm we study the fault prediction in fuzzy discrete event systems. Fuzzy discrete event systems are proposed to deal with vagueness, impreciseness, and subjectivity in real-world problems. The verification is divided into two steps. In the first step, we give a method to construct a Diagnoser. And in the second step, based on the structure of diagnoser we give the necessary and sufficient conditions to verify the future occurrence of the fault. The newly proposed approach allows us to deal with the problem of fault prediction for both crisp DESs and FDESs. Finally, an example is provided to illustrate the efficiency of the proposed approach. Subject Categories and Descriptors: [I.2.3 Deduction and Theorem Proving]; Uncertainty, “fuzzy,” and probabilistic reasoning: [B.1.3 Control Structure Reliability, Testing, and Fault-Tolerance] General Terms: Fault Prediction, Fuzzy Models, Discrete Events
{"title":"Fault Prediction in Fuzzy Discrete Event Systems: A Diagnoser Approach","authors":"Bilal Benmessahel, Farid Nouioua, Mohamed Touahria, A. Chariete","doi":"10.6025/jdim/2019/17/6/337-345","DOIUrl":"https://doi.org/10.6025/jdim/2019/17/6/337-345","url":null,"abstract":"In this workm we study the fault prediction in fuzzy discrete event systems. Fuzzy discrete event systems are proposed to deal with vagueness, impreciseness, and subjectivity in real-world problems. The verification is divided into two steps. In the first step, we give a method to construct a Diagnoser. And in the second step, based on the structure of diagnoser we give the necessary and sufficient conditions to verify the future occurrence of the fault. The newly proposed approach allows us to deal with the problem of fault prediction for both crisp DESs and FDESs. Finally, an example is provided to illustrate the efficiency of the proposed approach. Subject Categories and Descriptors: [I.2.3 Deduction and Theorem Proving]; Uncertainty, “fuzzy,” and probabilistic reasoning: [B.1.3 Control Structure Reliability, Testing, and Fault-Tolerance] General Terms: Fault Prediction, Fuzzy Models, Discrete Events","PeriodicalId":303976,"journal":{"name":"J. Digit. Inf. Manag.","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116463188","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 : 2019-12-01DOI: 10.6025/jdim/2019/17/6/313-320
Fabiano Tavares da Silva, J. Maia
The Semantic Distributional Model is based on the frequency of contexts of use of language terms in large open corpus such as the web, to establish similarity or the relationship between words. These relationships or similarities can be used to add terms when expanding queries. The idea explored in this paper is that, for queries in closed collections of text documents, a posterior filter based on the restricted vocabulary of the collection can improve the effectiveness of automatic query expansion. This idea is developed and evaluated in publicly available benchmarks presenting promising results. Subject Categories and Descriptors: [H.3.3 Information Search and Retrieval]; Query formulation: [I.2.7 Natural Language Processing] Text analysis [F.4.2 Grammars and Other Rewriting Systems]; Grammar types General Terms: Distributional Semantic Model, Information Retrieval, Local Context Analysis.
{"title":"Query Expansion in Text Information Retrieval with Local Context and Distributional Model","authors":"Fabiano Tavares da Silva, J. Maia","doi":"10.6025/jdim/2019/17/6/313-320","DOIUrl":"https://doi.org/10.6025/jdim/2019/17/6/313-320","url":null,"abstract":"The Semantic Distributional Model is based on the frequency of contexts of use of language terms in large open corpus such as the web, to establish similarity or the relationship between words. These relationships or similarities can be used to add terms when expanding queries. The idea explored in this paper is that, for queries in closed collections of text documents, a posterior filter based on the restricted vocabulary of the collection can improve the effectiveness of automatic query expansion. This idea is developed and evaluated in publicly available benchmarks presenting promising results. Subject Categories and Descriptors: [H.3.3 Information Search and Retrieval]; Query formulation: [I.2.7 Natural Language Processing] Text analysis [F.4.2 Grammars and Other Rewriting Systems]; Grammar types General Terms: Distributional Semantic Model, Information Retrieval, Local Context Analysis.","PeriodicalId":303976,"journal":{"name":"J. Digit. Inf. Manag.","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129530611","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 : 2019-10-01DOI: 10.6025/jdim/2019/17/5/270-288
Nadjib Mesbahi, Merouane Zoubeidi, Abdelhak Merizig, O. Kazar
{"title":"An Agent-Based Approach for Extracting Business Association Rules from Centralized Databases Systems","authors":"Nadjib Mesbahi, Merouane Zoubeidi, Abdelhak Merizig, O. Kazar","doi":"10.6025/jdim/2019/17/5/270-288","DOIUrl":"https://doi.org/10.6025/jdim/2019/17/5/270-288","url":null,"abstract":"","PeriodicalId":303976,"journal":{"name":"J. Digit. Inf. Manag.","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121939714","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 : 2019-10-01DOI: 10.6025/jdim/2019/17/5/289-295
Sadik Bessou, Rania Aberkane
This paper presents an approach based on supervised machine learning methods to discriminate between positive, negative and neutral Arabic reviews in online newswire. The corpus is labeled for subjectivity and sentiment analysis (SSA) at the sentence-level. The model uses both count and TF-IDF representations and apply six machine learning algorithms; Multinomial Naive Bayes, Support Vector Machines (SVM), Random Forest, Logistic Regression, Multi-layer perceptron and k-nearest neighbors using uni-grams, bi-grams features. With the goal of extracting users sentiment from written text. Experimental results showed that n-gram features could substantially improve performance; and showed that the Multinomial Naive Bayes approach is the most accurate in predicting topic polarity. Best results were achieved using count vectors trained by combination of word-based uni-grams and bi-grams with an overall accuracy of 85.57% over two classes and 65.64% over three classes.
{"title":"Subjective Sentiment Analysis for Arabic Newswire Comments","authors":"Sadik Bessou, Rania Aberkane","doi":"10.6025/jdim/2019/17/5/289-295","DOIUrl":"https://doi.org/10.6025/jdim/2019/17/5/289-295","url":null,"abstract":"This paper presents an approach based on supervised machine learning methods to discriminate between positive, negative and neutral Arabic reviews in online newswire. The corpus is labeled for subjectivity and sentiment analysis (SSA) at the sentence-level. The model uses both count and TF-IDF representations and apply six machine learning algorithms; Multinomial Naive Bayes, Support Vector Machines (SVM), Random Forest, Logistic Regression, Multi-layer perceptron and k-nearest neighbors using uni-grams, bi-grams features. With the goal of extracting users sentiment from written text. Experimental results showed that n-gram features could substantially improve performance; and showed that the Multinomial Naive Bayes approach is the most accurate in predicting topic polarity. Best results were achieved using count vectors trained by combination of word-based uni-grams and bi-grams with an overall accuracy of 85.57% over two classes and 65.64% over three classes.","PeriodicalId":303976,"journal":{"name":"J. Digit. Inf. Manag.","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131867349","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：