In recent years, we have witnessed a marked rise in the number of cloud service providers with each offering a plethora of cloud services with different objectives. Gaining confidence for cloud technology adoption as well as selecting a suitable cloud service provider, both require a proper evaluation of cloud service trust characteristics. Hence, the evaluation of cloud services before used by the customer is of utmost importance. In this article, we adapt the extracted trust characteristics from both system and software quality standards and cloud computing standards, for evaluating cloud services. Moreover, we derive measures for each trust characteristics to evaluate the trustworthiness of different cloud service providers, and generalize these trust measures for any type of cloud services (e.g. Software as a Service, Platform as a Service, and Infrastructure as a Service). Our work thereby demonstrates a way to apply generalized trust measures for cloud services and therefore contributes to a better understanding of cloud services to evaluate their quality characteristics. As part of our ongoing research, the results of this study will be used to develop a comprehensive cloud trust model.
{"title":"A Proposition of Modifications and Extensions of Cloud Computing Standards for Trust Characteristics Measures","authors":"S. Eftekhar, W. Suryn","doi":"10.5539/cis.v12n3p27","DOIUrl":"https://doi.org/10.5539/cis.v12n3p27","url":null,"abstract":"In recent years, we have witnessed a marked rise in the number of cloud service providers with each offering a plethora of cloud services with different objectives. Gaining confidence for cloud technology adoption as well as selecting a suitable cloud service provider, both require a proper evaluation of cloud service trust characteristics. Hence, the evaluation of cloud services before used by the customer is of utmost importance. In this article, we adapt the extracted trust characteristics from both system and software quality standards and cloud computing standards, for evaluating cloud services. Moreover, we derive measures for each trust characteristics to evaluate the trustworthiness of different cloud service providers, and generalize these trust measures for any type of cloud services (e.g. Software as a Service, Platform as a Service, and Infrastructure as a Service). Our work thereby demonstrates a way to apply generalized trust measures for cloud services and therefore contributes to a better understanding of cloud services to evaluate their quality characteristics. As part of our ongoing research, the results of this study will be used to develop a comprehensive cloud trust model.","PeriodicalId":14676,"journal":{"name":"J. Chem. Inf. Comput. Sci.","volume":"132 31","pages":"27-41"},"PeriodicalIF":0.0,"publicationDate":"2019-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91433095","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}
Drawing on personality traits theory (Costa & McCrae, 1985) and organizational commitment theory (Mowday, Steers & Porter, 1979), the purpose of the present study was to investigate, through four separate hierarchical regression procedures, the effect of a set of independent variables (neuroticism, gender, and generational age) on four separate dependent variables (DV): overall organizational commitment (OC), affective commitment (AC), continuance commitment (CC), and normative commitment (NC). The sample consisted of responses from 279 IT professionals in the United States, drawn from a national sample from the merged cross-sectional GSS 1972-2014 Cross-Sectional Cumulative Data, Release 5, March 24, 2016. Results of multiple regressions analyses revealed that, among IT professionals, neuroticism did not predict overall OC, AC, CC, or NC. Generational age predicted OC, AC, and CC with statistical significance. Gender predicted CC; none of the independent variables (IVs) predicted NC. Directions for future research are offered.
{"title":"IT Professionals' Personality, Personal Characteristics, and Commitment: Evidence from a National Survey","authors":"J. Syed, Ruth M. Tappin","doi":"10.5539/cis.v12n3p58","DOIUrl":"https://doi.org/10.5539/cis.v12n3p58","url":null,"abstract":"Drawing on personality traits theory (Costa & McCrae, 1985) and organizational commitment theory (Mowday, Steers & Porter, 1979), the purpose of the present study was to investigate, through four separate hierarchical regression procedures, the effect of a set of independent variables (neuroticism, gender, and generational age) on four separate dependent variables (DV): overall organizational commitment (OC), affective commitment (AC), continuance commitment (CC), and normative commitment (NC). The sample consisted of responses from 279 IT professionals in the United States, drawn from a national sample from the merged cross-sectional GSS 1972-2014 Cross-Sectional Cumulative Data, Release 5, March 24, 2016. Results of multiple regressions analyses revealed that, among IT professionals, neuroticism did not predict overall OC, AC, CC, or NC. Generational age predicted OC, AC, and CC with statistical significance. Gender predicted CC; none of the independent variables (IVs) predicted NC. Directions for future research are offered.","PeriodicalId":14676,"journal":{"name":"J. Chem. Inf. Comput. Sci.","volume":"40 1","pages":"58-71"},"PeriodicalIF":0.0,"publicationDate":"2019-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87179791","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}
Driss El Hannach, R. Marghoubi, Z. E. Akkaoui, Mohamed Dahchour
The paramount importance of project portfolios for business drives managers to search for highly efficient support tools to overcome complex challenges of their management. A major tradeoff is to acquire tools able to produce a convenient portfolio project prioritization process, on which business investments are decided. However, by using existing Project Portfolio Management Systems (PPMS), many concurrent projects in a portfolio are usually prioritized and planned in the upstream life-cycle phases according to financial criteria, and overlooking the portfolio alignment to enterprise strategies and the availability of resources, although their importance. In this paper, we propose a conceptual formalization of PPMS with respect to a double portfolio prioritization process that performs two levels of selections according to both: i.) Strategy alignment, including returns on investment, size, and total cost; and ii.) Execution capability, as the organization should be able to manage and deliver the selected projects' outcomes. The advantage of our PPMS framework is twofold. First, it is useful to be customized by designers to fit organization needs. Second it is built with respect to the double prioritization phase process, as an end-to-end process that guarantees optimal portfolios generation. Further, the proposed PPMS system and its identified functionalities are validated through an implementation of a prototype tool.
{"title":"Analysis and Design of a Project Portfolio Management System","authors":"Driss El Hannach, R. Marghoubi, Z. E. Akkaoui, Mohamed Dahchour","doi":"10.5539/cis.v12n3p42","DOIUrl":"https://doi.org/10.5539/cis.v12n3p42","url":null,"abstract":"The paramount importance of project portfolios for business drives managers to search for highly efficient support tools to overcome complex challenges of their management. A major tradeoff is to acquire tools able to produce a convenient portfolio project prioritization process, on which business investments are decided. However, by using existing Project Portfolio Management Systems (PPMS), many concurrent projects in a portfolio are usually prioritized and planned in the upstream life-cycle phases according to financial criteria, and overlooking the portfolio alignment to enterprise strategies and the availability of resources, although their importance. In this paper, we propose a conceptual formalization of PPMS with respect to a double portfolio prioritization process that performs two levels of selections according to both: i.) Strategy alignment, including returns on investment, size, and total cost; and ii.) Execution capability, as the organization should be able to manage and deliver the selected projects' outcomes. The advantage of our PPMS framework is twofold. First, it is useful to be customized by designers to fit organization needs. Second it is built with respect to the double prioritization phase process, as an end-to-end process that guarantees optimal portfolios generation. Further, the proposed PPMS system and its identified functionalities are validated through an implementation of a prototype tool.","PeriodicalId":14676,"journal":{"name":"J. Chem. Inf. Comput. Sci.","volume":"25 1","pages":"42-57"},"PeriodicalIF":0.0,"publicationDate":"2019-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87658512","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}
Soumaya Amraoui, Mina Elmaallam, Hicham Bensaid, A. Kriouile
The security of a company's information system (IS) is an important requirement for the pursuit of its business. Risk management contributes to the protection of the IS assets. It saves the organism from the losses caused by the emergence of unwanted events having an incidence on the IS objectives and consequently on its strategy. It has also an important role in the decision making about entering new opportunities. In addition, it promises an optimal allocation of information system resources. The risk management process aims to analyze what can happen and what are the eventual consequences for the organization before deciding what needs to be done and reducing the risks to an acceptable level. This paper presents a literature review of IS risk management and gives a comparative analyse of its processes, methods and standards.
{"title":"Information Systems Risk Management: Litterature Review","authors":"Soumaya Amraoui, Mina Elmaallam, Hicham Bensaid, A. Kriouile","doi":"10.5539/CIS.V12N3P1","DOIUrl":"https://doi.org/10.5539/CIS.V12N3P1","url":null,"abstract":"The security of a company's information system (IS) is an important requirement for the pursuit of its business. Risk management contributes to the protection of the IS assets. It saves the organism from the losses caused by the emergence of unwanted events having an incidence on the IS objectives and consequently on its strategy. It has also an important role in the decision making about entering new opportunities. In addition, it promises an optimal allocation of information system resources. The risk management process aims to analyze what can happen and what are the eventual consequences for the organization before deciding what needs to be done and reducing the risks to an acceptable level. This paper presents a literature review of IS risk management and gives a comparative analyse of its processes, methods and standards.","PeriodicalId":14676,"journal":{"name":"J. Chem. Inf. Comput. Sci.","volume":"72 1","pages":"1-20"},"PeriodicalIF":0.0,"publicationDate":"2019-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90896803","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}
Complexity of Arabic writing language makes its handwritten recognition very complex in terms of computer algorithms. The Arabic handwritten recognition has high importance in modern applications. The contour analysis of word image can extract special contour features that discriminate one character from another by the mean of vector features. This paper implements a set of pre-processing functions over a handwritten Arabic characters, with contour analysis, to enter the contour vector to neural network to recognize it. The selection of this set of pre-processing algorithms was completed after hundreds of tests and validation. The feed forward neural network architecture was trained using many patterns regardless of the Arabic font style building a rigid recognition model. Because of the shortcomings in Arabic written databases or datasets, the testing was done by non-standard data set. The presented algorithm structure got recognition ratio about 97%.
{"title":"Arabic Hand Written Character Recognition Based on Contour Matching and Neural Network","authors":"Marwan Abo Zanona, A. Abuhamdah, B. El-Zaghmouri","doi":"10.5539/CIS.V12N2P126","DOIUrl":"https://doi.org/10.5539/CIS.V12N2P126","url":null,"abstract":"Complexity of Arabic writing language makes its handwritten recognition very complex in terms of computer algorithms. The Arabic handwritten recognition has high importance in modern applications. The contour analysis of word image can extract special contour features that discriminate one character from another by the mean of vector features. This paper implements a set of pre-processing functions over a handwritten Arabic characters, with contour analysis, to enter the contour vector to neural network to recognize it. The selection of this set of pre-processing algorithms was completed after hundreds of tests and validation. The feed forward neural network architecture was trained using many patterns regardless of the Arabic font style building a rigid recognition model. Because of the shortcomings in Arabic written databases or datasets, the testing was done by non-standard data set. The presented algorithm structure got recognition ratio about 97%.","PeriodicalId":14676,"journal":{"name":"J. Chem. Inf. Comput. Sci.","volume":"18 1","pages":"126-137"},"PeriodicalIF":0.0,"publicationDate":"2019-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74113356","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 large percentage of cancer patients are breast cancer patients. The main available methodology to examine the breast cancer is the Mammography. It detects the signs of breast cancer as different signs supports the experts’ decision. Actually, the Mammography is based on human perception and observations. So, build an AI computerized system will take major role in early signs detection. This paper presents an image processing with aid of artificial neural networks computations for computerized signs detection and exploration of breast cancer. The input material is the mammogram images, and the output helps the pathologists to take a decision. A set of input mammogram images was used for development, testing, and evaluation. The mammographic image will be preprocessed and then the features will be extracted using discrete wavelet transformation with aid of Weiner filtration. A historical data of extracted features were used to train a neural network, while the historical extracted features contains both Cancer and non-Cancer images. The combination of neural network machine learning, and rigid image processing techniques resulted accurate outputs. The methodology and results are showed and discussed later in this paper.
{"title":"Image Processing & Neural Network Based Breast Cancer Detection","authors":"Marwan Abo Zanona","doi":"10.5539/CIS.V12N2P146","DOIUrl":"https://doi.org/10.5539/CIS.V12N2P146","url":null,"abstract":"A large percentage of cancer patients are breast cancer patients. The main available methodology to examine the breast cancer is the Mammography. It detects the signs of breast cancer as different signs supports the experts’ decision. Actually, the Mammography is based on human perception and observations. So, build an AI computerized system will take major role in early signs detection. This paper presents an image processing with aid of artificial neural networks computations for computerized signs detection and exploration of breast cancer. The input material is the mammogram images, and the output helps the pathologists to take a decision. A set of input mammogram images was used for development, testing, and evaluation. The mammographic image will be preprocessed and then the features will be extracted using discrete wavelet transformation with aid of Weiner filtration. A historical data of extracted features were used to train a neural network, while the historical extracted features contains both Cancer and non-Cancer images. The combination of neural network machine learning, and rigid image processing techniques resulted accurate outputs. The methodology and results are showed and discussed later in this paper.","PeriodicalId":14676,"journal":{"name":"J. Chem. Inf. Comput. Sci.","volume":"8 1","pages":"146-"},"PeriodicalIF":0.0,"publicationDate":"2019-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90323121","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}
Software Product Lines represent a solution for massive development with minimum costs, while assuring product high quality and interesting time to market. In fact, Software Product Lines systems are used for massive productions, and are based on systematic reuse of commun components, while offering the ability to add specific development, in order to satisfy particular users or market needs. However, to maintain such complex and large-scale systems, it is mandatory to adopt a suitable tracing policy that satisfies the system constraints, especially cost and complexity. Unfortunately, tracing is rearly applied in Software Product Lines as it presents several constraints, especially its cost. Through our research work, we tried to come up with elements that would help break this prejudice. Therefore, we worked on a cost and Return on Investment estimation model that helps identify the optimal conditions (phase and policy) for implementing a tracing solution. As a result of our work, we found that implementing specific trace links, in a targeted approach that meets business goals, and starting from the Domain Engineering phase, costs less and presents the most interesting Return on Investment. To conduct this study and reach those findings, we followed the Design Science Research Methodology. In this article, we detail the steps of our research according to this methodology’s phases.
{"title":"Towards Efficient Tracing in Software Product Lines: Research Methodology","authors":"Zineb Mcharfi, B. E. Asri","doi":"10.5539/CIS.V12N2P138","DOIUrl":"https://doi.org/10.5539/CIS.V12N2P138","url":null,"abstract":"Software Product Lines represent a solution for massive development with minimum costs, while assuring product high quality and interesting time to market. In fact, Software Product Lines systems are used for massive productions, and are based on systematic reuse of commun components, while offering the ability to add specific development, in order to satisfy particular users or market needs. However, to maintain such complex and large-scale systems, it is mandatory to adopt a suitable tracing policy that satisfies the system constraints, especially cost and complexity. Unfortunately, tracing is rearly applied in Software Product Lines as it presents several constraints, especially its cost. Through our research work, we tried to come up with elements that would help break this prejudice. Therefore, we worked on a cost and Return on Investment estimation model that helps identify the optimal conditions (phase and policy) for implementing a tracing solution. As a result of our work, we found that implementing specific trace links, in a targeted approach that meets business goals, and starting from the Domain Engineering phase, costs less and presents the most interesting Return on Investment. To conduct this study and reach those findings, we followed the Design Science Research Methodology. In this article, we detail the steps of our research according to this methodology’s phases.","PeriodicalId":14676,"journal":{"name":"J. Chem. Inf. Comput. Sci.","volume":"173 1","pages":"138-145"},"PeriodicalIF":0.0,"publicationDate":"2019-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79580721","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}
Udoinyang Godwin Inyang, U. Umoh, Ifeoma Nnaemeka, Samuel A. Robinson
The large nature of students’ dataset has made it difficult to find patterns associated with students’ academic performance (AP) using conventional methods. This has increased the rate of drop-outs, graduands with weak class of degree (CoD) and students that spend more than the minimum stipulated duration of studies. It is necessary to determine students’ AP using educational data mining (EDM) tools in order to know students who are likely to perform poorly at an early stage of their studies. This paper explores k-means and self-organizing map (SOM) in mining pieces of knowledge relating to the natural number of clusters in students’ dataset and the association of the input features using selected demographic, pre-admission and first year performance. Matlab 2015a was the programming environment and the dataset consists of nine sets of computer science graduands. Cluster validity assessment with k-means discovered four (4) clusters with correlation metric yielding the highest mean silhouette value of 0.5912. SOM provided an hexagonal grid visual of feature component planes and scatter plots of each significant input attribute. The result shows that the significant attributes were highly correlated with each other except entry mode (EM), indicating that the impact of EM on CoD varies with students irrespective of mode of admission. Also, four distinct clusters were also discovered in the dataset by SOM —7.7% belonging to cluster 1 (first class), and 25% for cluster 2 (2nd class Upper) while Clusters 3 and 4 had 35% proportion each. This validates the results of k-means and further confirms the importance of early detection of students’ AP and confirms the effectiveness of SOM as a cluster validity tool. As further work, the labels from SOM will be associated with records in the dataset for association rule mining, supervised learning and prediction of students’ AP.
{"title":"Unsupervised Characterization and Visualization of Students' Academic Performance Features","authors":"Udoinyang Godwin Inyang, U. Umoh, Ifeoma Nnaemeka, Samuel A. Robinson","doi":"10.5539/CIS.V12N2P103","DOIUrl":"https://doi.org/10.5539/CIS.V12N2P103","url":null,"abstract":"The large nature of students’ dataset has made it difficult to find patterns associated with students’ academic performance (AP) using conventional methods. This has increased the rate of drop-outs, graduands with weak class of degree (CoD) and students that spend more than the minimum stipulated duration of studies. It is necessary to determine students’ AP using educational data mining (EDM) tools in order to know students who are likely to perform poorly at an early stage of their studies. This paper explores k-means and self-organizing map (SOM) in mining pieces of knowledge relating to the natural number of clusters in students’ dataset and the association of the input features using selected demographic, pre-admission and first year performance. Matlab 2015a was the programming environment and the dataset consists of nine sets of computer science graduands. Cluster validity assessment with k-means discovered four (4) clusters with correlation metric yielding the highest mean silhouette value of 0.5912. SOM provided an hexagonal grid visual of feature component planes and scatter plots of each significant input attribute. The result shows that the significant attributes were highly correlated with each other except entry mode (EM), indicating that the impact of EM on CoD varies with students irrespective of mode of admission. Also, four distinct clusters were also discovered in the dataset by SOM —7.7% belonging to cluster 1 (first class), and 25% for cluster 2 (2nd class Upper) while Clusters 3 and 4 had 35% proportion each. This validates the results of k-means and further confirms the importance of early detection of students’ AP and confirms the effectiveness of SOM as a cluster validity tool. As further work, the labels from SOM will be associated with records in the dataset for association rule mining, supervised learning and prediction of students’ AP.","PeriodicalId":14676,"journal":{"name":"J. Chem. Inf. Comput. Sci.","volume":"98 1","pages":"103-116"},"PeriodicalIF":0.0,"publicationDate":"2019-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77343156","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}
Background: Information system use has substantially increased among the organization based on its effective integration of the resources and improved performance. The increasing reliance on the information system serves as a great security threat for the firms. Objective: The study intends to evaluate the security of the information system in the organization located in the region of Saudi Arabia, concerning the user’s awareness level. Methods: The quantitative design of the study is adopted which uses the survey approach. A close-ended questionnaire is used for evaluating the awareness level among the individuals. A total of 109 participants (males and females) in the Saudi Company were recruited for the study. Results: Despite the implementation of the policy, employees were unaware of it. The study highlights that the development of the firm’s information security policy requires the firm to make employees aware of the significance of the information security. Conclusion: The study concludes that the organization needs to educate the workforce of the information security policy and develop their necessary understanding of the information security system. This allows the employees to identify and report security threats and risks which helps in the improvement of information security awareness.
{"title":"Safeguarding the Information Systems in an Organization through Different Technologies, Policies, and Actions","authors":"H. Alkahtani","doi":"10.5539/CIS.V12N2P117","DOIUrl":"https://doi.org/10.5539/CIS.V12N2P117","url":null,"abstract":"Background: Information system use has substantially increased among the organization based on its effective integration of the resources and improved performance. The increasing reliance on the information system serves as a great security threat for the firms. Objective: The study intends to evaluate the security of the information system in the organization located in the region of Saudi Arabia, concerning the user’s awareness level. Methods: The quantitative design of the study is adopted which uses the survey approach. A close-ended questionnaire is used for evaluating the awareness level among the individuals. A total of 109 participants (males and females) in the Saudi Company were recruited for the study. Results: Despite the implementation of the policy, employees were unaware of it. The study highlights that the development of the firm’s information security policy requires the firm to make employees aware of the significance of the information security. Conclusion: The study concludes that the organization needs to educate the workforce of the information security policy and develop their necessary understanding of the information security system. This allows the employees to identify and report security threats and risks which helps in the improvement of information security awareness.","PeriodicalId":14676,"journal":{"name":"J. Chem. Inf. Comput. Sci.","volume":"253 1","pages":"117-125"},"PeriodicalIF":0.0,"publicationDate":"2019-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74998487","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}
Online reviews play an increasingly important role in the purchase decisions of potential customers. Incidentally, driven by the desire to gain profit or publicity, spammers may be hired to write fake reviews and promote or demote the reputation of products or services. Correspondingly, opinion spam detection has attracted attention from both business and research communities in recent years. However, unlike other tasks such as news classification or blog classification, the existing review spam datasets are typically limited due to the expensiveness of human annotation, which may further affect detection performance even if excellent classifiers have been developed. We propose a novel approach in this paper to boost opinion spam detection performance by fully utilizing the existing labelled small-size dataset. We first design an annotation extension scheme that uses extra tree classifiers to train multiple estimators and then iteratively generate reliable labelled samples from unlabeled ones. Subsequently, we examine neural network scenarios on a newly extended dataset to learn the distributed representation. Experimental results suggest that the proposed approach has better generalization capability and improved performance than state-of-the-art methods.
{"title":"Opinion Spam Detection based on Annotation Extension and Neural Networks","authors":"Yuanchao Liu, B. Pang","doi":"10.5539/CIS.V12N2P87","DOIUrl":"https://doi.org/10.5539/CIS.V12N2P87","url":null,"abstract":"Online reviews play an increasingly important role in the purchase decisions of potential customers. Incidentally, driven by the desire to gain profit or publicity, spammers may be hired to write fake reviews and promote or demote the reputation of products or services. Correspondingly, opinion spam detection has attracted attention from both business and research communities in recent years. However, unlike other tasks such as news classification or blog classification, the existing review spam datasets are typically limited due to the expensiveness of human annotation, which may further affect detection performance even if excellent classifiers have been developed. We propose a novel approach in this paper to boost opinion spam detection performance by fully utilizing the existing labelled small-size dataset. We first design an annotation extension scheme that uses extra tree classifiers to train multiple estimators and then iteratively generate reliable labelled samples from unlabeled ones. Subsequently, we examine neural network scenarios on a newly extended dataset to learn the distributed representation. Experimental results suggest that the proposed approach has better generalization capability and improved performance than state-of-the-art methods.","PeriodicalId":14676,"journal":{"name":"J. Chem. Inf. Comput. Sci.","volume":"45 1","pages":"87-"},"PeriodicalIF":0.0,"publicationDate":"2019-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82812592","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：