Recommender Systems (RS) seek to present resources with features that meet users' interest and needs. The use of these systems with social networks, which have characteristics of information propagation and sharing, enables the dissemination of recommended resources. In order to study and identify the information diffusion metrics that are most related to social networks, this article presents an architecture, with the purpose of recommending educational resources to diffusing people, increasing the social network reach. The proposal was evaluated through a feasibility study and a case study in an educational context on the Twitter social network. The results point to the feasibility of the proposal, not restricted to its validation context.
{"title":"Information Diffusion in Social Networks: a recommendation model in the educational context","authors":"Felipe Neves, Victor Ströele, Fernanda Campos","doi":"10.1145/3330204.3330234","DOIUrl":"https://doi.org/10.1145/3330204.3330234","url":null,"abstract":"Recommender Systems (RS) seek to present resources with features that meet users' interest and needs. The use of these systems with social networks, which have characteristics of information propagation and sharing, enables the dissemination of recommended resources. In order to study and identify the information diffusion metrics that are most related to social networks, this article presents an architecture, with the purpose of recommending educational resources to diffusing people, increasing the social network reach. The proposal was evaluated through a feasibility study and a case study in an educational context on the Twitter social network. The results point to the feasibility of the proposal, not restricted to its validation context.","PeriodicalId":348938,"journal":{"name":"Proceedings of the XV Brazilian Symposium on Information Systems","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131697068","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}
Rafael Pereira de Oliveira, F. Baião, Ana Carolina Almeida, D. Schwabe, Sérgio Lifschitz
Database tuning is a crucial task to address the performance of information systems that deal with a considerable amount of information stored in databases. Current tuning tools are very platform-specific and do not provide adequate support for the database administrator to reason about performance improvement suggestions. In this paper, we discuss several architectural and implementation decisions of Outer-Tuning, our framework that supports database tuning. Outer-Tuning follows a model-driven development and a modular architecture design, which enabled several benefits. This paper contributes with: (i) the architectural design model adopted in Outer-Tuning, which combines imperative and declarative programming; (ii) the discussions and steps to integrate several software components; and (iii) the actual framework implementation. We assess our framework with an experiment using the TPC-H benchmark. The results evidence that Outer-Tuning infers useful tuning actions and supports the DBA by providing a more semantic environment to create and adapt tuning heuristics using concepts closer to his/her domain, and also relevant information on the rationale of the tuning actions through a friendly web interface.
{"title":"Outer-Tuning: an integration of rules, ontology and RDBMS","authors":"Rafael Pereira de Oliveira, F. Baião, Ana Carolina Almeida, D. Schwabe, Sérgio Lifschitz","doi":"10.1145/3330204.3330270","DOIUrl":"https://doi.org/10.1145/3330204.3330270","url":null,"abstract":"Database tuning is a crucial task to address the performance of information systems that deal with a considerable amount of information stored in databases. Current tuning tools are very platform-specific and do not provide adequate support for the database administrator to reason about performance improvement suggestions. In this paper, we discuss several architectural and implementation decisions of Outer-Tuning, our framework that supports database tuning. Outer-Tuning follows a model-driven development and a modular architecture design, which enabled several benefits. This paper contributes with: (i) the architectural design model adopted in Outer-Tuning, which combines imperative and declarative programming; (ii) the discussions and steps to integrate several software components; and (iii) the actual framework implementation. We assess our framework with an experiment using the TPC-H benchmark. The results evidence that Outer-Tuning infers useful tuning actions and supports the DBA by providing a more semantic environment to create and adapt tuning heuristics using concepts closer to his/her domain, and also relevant information on the rationale of the tuning actions through a friendly web interface.","PeriodicalId":348938,"journal":{"name":"Proceedings of the XV Brazilian Symposium on Information Systems","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114210253","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}
José S. Barroso Júnior, M. Pimentel, V. Nunes, C. Cappelli
The growing demand for information about politicians and the speed with which news is propagated by media and social networks reveals in contemporary times a more participatory view of the citizen in politics and social control in government actions. Speculation about information about politicians and their respective parties have become more constant, such as in electoral periods, investigations on processes, journalistic interest, etc. This can be observed during the electoral period of 2018 in which the search for information about the candidates and their respective parties was notorious. Given that information transparency is paramount for a democratic regime, the organization and consolidation of data from official sources is a reliable tool for consultation and dissemination of knowledge. According to this period, this work sought to understand some cognitive models that explain electoral behavior, which led to the investigation of factors that influence the decision to vote. The observed cognitive models indicate some attitudes, opinions, satisfactions, events and government assessments that interfere to some degree in the choice of vote, approval or disapproval for some purpose. Among the factors of influence observed, there are prosopographic information, which deals with the biography of politicians, and which are available in the various transparency portals. However, the data contained in these platforms are not organized in such a way as to make the information more intelligible and more reliable to the citizen. In this sense, this research is interested in the prosopographic information content, which would be better evaluated if the data of the politicians were organized according to a conceptual model of knowledge. The methodological approach adopted in this research is the Design Science Reserch (DSR), which directs the construction of an artifact in a given context, whose theoretical conjectures are based on the search and production of knowledge. As a result, research contributes to the knowledge base, as it discusses the evolution of cognitive models and through the design of the artifact, delivers reliable and relevant information about the life of a politician.
{"title":"Design Science Research to design a conceptual model about prosopographic information related to politicians","authors":"José S. Barroso Júnior, M. Pimentel, V. Nunes, C. Cappelli","doi":"10.1145/3330204.3330233","DOIUrl":"https://doi.org/10.1145/3330204.3330233","url":null,"abstract":"The growing demand for information about politicians and the speed with which news is propagated by media and social networks reveals in contemporary times a more participatory view of the citizen in politics and social control in government actions. Speculation about information about politicians and their respective parties have become more constant, such as in electoral periods, investigations on processes, journalistic interest, etc. This can be observed during the electoral period of 2018 in which the search for information about the candidates and their respective parties was notorious. Given that information transparency is paramount for a democratic regime, the organization and consolidation of data from official sources is a reliable tool for consultation and dissemination of knowledge. According to this period, this work sought to understand some cognitive models that explain electoral behavior, which led to the investigation of factors that influence the decision to vote. The observed cognitive models indicate some attitudes, opinions, satisfactions, events and government assessments that interfere to some degree in the choice of vote, approval or disapproval for some purpose. Among the factors of influence observed, there are prosopographic information, which deals with the biography of politicians, and which are available in the various transparency portals. However, the data contained in these platforms are not organized in such a way as to make the information more intelligible and more reliable to the citizen. In this sense, this research is interested in the prosopographic information content, which would be better evaluated if the data of the politicians were organized according to a conceptual model of knowledge. The methodological approach adopted in this research is the Design Science Reserch (DSR), which directs the construction of an artifact in a given context, whose theoretical conjectures are based on the search and production of knowledge. As a result, research contributes to the knowledge base, as it discusses the evolution of cognitive models and through the design of the artifact, delivers reliable and relevant information about the life of a politician.","PeriodicalId":348938,"journal":{"name":"Proceedings of the XV Brazilian Symposium on Information Systems","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124350506","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}
Denisson Santana dos Santos, H. Nunes, Hendrik T. Macedo, A. C. Neto
Mobile application developers, aimming to make their applications available to the largest number of platforms, are increasingly seeking to optimize how to develop mobile applications. Creating the same specific application for each platform demands greater financial, time and labor investment. In order to reduce these problems, mobile development frameworks for multiplatforms have emerged, making possible to create multi-platform applications from the same source code. However, a new issue arises for the developer - the plurality and diversity of these frameworks, each one with objectives and characteristics that sometimes make them closer and in other cases farther. In addition, there is a demand for multi-purpose applications and work teams with different training profiles. For these reasons, choosing a framework is not trivial. This paper presents an approach that indicates which multi-platform application development framework is most suitable for a given project, implemented as a knowledge-based recommendation system that considered the critical factors of a given project.
{"title":"Recommendation System for Cross-Platform Mobile Development Framework","authors":"Denisson Santana dos Santos, H. Nunes, Hendrik T. Macedo, A. C. Neto","doi":"10.1145/3330204.3330279","DOIUrl":"https://doi.org/10.1145/3330204.3330279","url":null,"abstract":"Mobile application developers, aimming to make their applications available to the largest number of platforms, are increasingly seeking to optimize how to develop mobile applications. Creating the same specific application for each platform demands greater financial, time and labor investment. In order to reduce these problems, mobile development frameworks for multiplatforms have emerged, making possible to create multi-platform applications from the same source code. However, a new issue arises for the developer - the plurality and diversity of these frameworks, each one with objectives and characteristics that sometimes make them closer and in other cases farther. In addition, there is a demand for multi-purpose applications and work teams with different training profiles. For these reasons, choosing a framework is not trivial. This paper presents an approach that indicates which multi-platform application development framework is most suitable for a given project, implemented as a knowledge-based recommendation system that considered the critical factors of a given project.","PeriodicalId":348938,"journal":{"name":"Proceedings of the XV Brazilian Symposium on Information Systems","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122855825","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 Software Product Line (SPL) consists of a systematic reuse strategy to construct systems with less effort as long as they belong to the same family that share the same components and belong to the same domain of Marketplace. In this context, to support large-scale reuse, components of a Software Product Line should be easy to maintain. Thus, developers should be more concerned with anomalies known as code smells and more than that, co-occurrences known as Inter-smell deserve to be further studied to verify their real impact on maintainability in SPL. Thus, this paper conducts a study to investigate the impact of Inter-smell occurrences on maintainability in MobileMedia and Health Watcher SPLs. The results show that the presence of co-occurrences of Inter-smell did not negatively impact the maintenance of MobileMedia and Health Watcher SPLs, unlike results found in other studies in the literature, and even more, our results indicate that the metric Lack of Cohesion of Methods is one of the most important for the maintainability of object-oriented SPLs.
{"title":"Analyzing the Impact of Inter-smell Relations on Software Maintainability: An Empirical Study with Software Product Lines","authors":"Júlio Martins, C. Bezerra, Anderson G. Uchôa","doi":"10.1145/3330204.3330254","DOIUrl":"https://doi.org/10.1145/3330204.3330254","url":null,"abstract":"A Software Product Line (SPL) consists of a systematic reuse strategy to construct systems with less effort as long as they belong to the same family that share the same components and belong to the same domain of Marketplace. In this context, to support large-scale reuse, components of a Software Product Line should be easy to maintain. Thus, developers should be more concerned with anomalies known as code smells and more than that, co-occurrences known as Inter-smell deserve to be further studied to verify their real impact on maintainability in SPL. Thus, this paper conducts a study to investigate the impact of Inter-smell occurrences on maintainability in MobileMedia and Health Watcher SPLs. The results show that the presence of co-occurrences of Inter-smell did not negatively impact the maintenance of MobileMedia and Health Watcher SPLs, unlike results found in other studies in the literature, and even more, our results indicate that the metric Lack of Cohesion of Methods is one of the most important for the maintainability of object-oriented SPLs.","PeriodicalId":348938,"journal":{"name":"Proceedings of the XV Brazilian Symposium on Information Systems","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126572850","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}
Douglas F. Pereira, P. Bugatti, Fabricio M. Lopes, A. L. Souza, P. T. Saito
This work presents an approach for the automatic processing and classification of the vigor of soybean seeds. The validation was performed through experimental evaluation using real datasets. By automating the seed analysis, it is possible a faster and more precise control, increasing the market competitiveness between producers and benefiting consumers with higher quality products and more appropriate and specific prices, according to the quality of the seeds. In addition, consequently, improvements related to sustainability and productivity can be obtained in the processes of planting, multiplication and/or commercialization of soybeans with better quality.
{"title":"An Intelligent System to Enhance the Productivity and Sustainability in Soybean Crop Enterprises","authors":"Douglas F. Pereira, P. Bugatti, Fabricio M. Lopes, A. L. Souza, P. T. Saito","doi":"10.1145/3330204.3330266","DOIUrl":"https://doi.org/10.1145/3330204.3330266","url":null,"abstract":"This work presents an approach for the automatic processing and classification of the vigor of soybean seeds. The validation was performed through experimental evaluation using real datasets. By automating the seed analysis, it is possible a faster and more precise control, increasing the market competitiveness between producers and benefiting consumers with higher quality products and more appropriate and specific prices, according to the quality of the seeds. In addition, consequently, improvements related to sustainability and productivity can be obtained in the processes of planting, multiplication and/or commercialization of soybeans with better quality.","PeriodicalId":348938,"journal":{"name":"Proceedings of the XV Brazilian Symposium on Information Systems","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132122815","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}
Mayrton D. de Queiroz, Ruan A. P. Palmeira, Felipe T. de Melo, Rodrigo G. Daniel, Ícaro T. de A. Rique, A. Guimarães, Marcelle Batista Martins, N. Lino
In the context of Smart Cities, there are concerns regarding Urban Mobility, which consists of identifying alternatives for the reduction of traffic of individual vehicles, better occupation of urban space, among other aspects. An alternative is the adoption of electric trains. However, a problem concerning energy consumption arises. Thus, this work aims to propose a framework based on Genetic Algorithms (GAs), called SmartSubway, to assist specialists with insertion of domain information in the problem of energy efficiency in electric trains in order to identify energy efficient driving profiles. As proof of concept, a system inspired in GAs was implemented. To validate the system, the domain information of a real scenario was inserted, where it was possible to carry out six experiments and identify the ones that obtained the best results.
{"title":"A Framework to Support Experts in the Study of Energy Efficiency in Urban Trains","authors":"Mayrton D. de Queiroz, Ruan A. P. Palmeira, Felipe T. de Melo, Rodrigo G. Daniel, Ícaro T. de A. Rique, A. Guimarães, Marcelle Batista Martins, N. Lino","doi":"10.1145/3330204.3330214","DOIUrl":"https://doi.org/10.1145/3330204.3330214","url":null,"abstract":"In the context of Smart Cities, there are concerns regarding Urban Mobility, which consists of identifying alternatives for the reduction of traffic of individual vehicles, better occupation of urban space, among other aspects. An alternative is the adoption of electric trains. However, a problem concerning energy consumption arises. Thus, this work aims to propose a framework based on Genetic Algorithms (GAs), called SmartSubway, to assist specialists with insertion of domain information in the problem of energy efficiency in electric trains in order to identify energy efficient driving profiles. As proof of concept, a system inspired in GAs was implemented. To validate the system, the domain information of a real scenario was inserted, where it was possible to carry out six experiments and identify the ones that obtained the best results.","PeriodicalId":348938,"journal":{"name":"Proceedings of the XV Brazilian Symposium on Information Systems","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132160252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In recent years, Deep Learning has presented impressive performance when solving complex image classification and recognition tasks in decision support systems. Nonetheless, studies have demonstrated Deep Learning models are susceptible to attacks conducted with adversarial images, i.e. images containing subtle perturbations in order to induce models to misclassification. The main existing countermeasures against adversarial images have shown ineficiency, permitting attackers to map their internal operation more easily. Therefore, this work aims to evaluate a defense method called MultiMagNet which randomly incorporates at runtime multiple defense components, implemented as autoencoders, in order to introduce an expanded form of non-determinism behavior for hindering evasions of adversarial nature. Experiments on CIFAR-10 dataset showed MultiMagNet was able to detect images generated by different attack algorithms.
{"title":"A Non-Deterministic Method to Construct Ensemble-Based Classifiers to Protect Decision Support Systems Against Adversarial Images: A Case Study","authors":"G. R. Machado, Eugênio Silva, R. Goldschmidt","doi":"10.1145/3330204.3330282","DOIUrl":"https://doi.org/10.1145/3330204.3330282","url":null,"abstract":"In recent years, Deep Learning has presented impressive performance when solving complex image classification and recognition tasks in decision support systems. Nonetheless, studies have demonstrated Deep Learning models are susceptible to attacks conducted with adversarial images, i.e. images containing subtle perturbations in order to induce models to misclassification. The main existing countermeasures against adversarial images have shown ineficiency, permitting attackers to map their internal operation more easily. Therefore, this work aims to evaluate a defense method called MultiMagNet which randomly incorporates at runtime multiple defense components, implemented as autoencoders, in order to introduce an expanded form of non-determinism behavior for hindering evasions of adversarial nature. Experiments on CIFAR-10 dataset showed MultiMagNet was able to detect images generated by different attack algorithms.","PeriodicalId":348938,"journal":{"name":"Proceedings of the XV Brazilian Symposium on Information Systems","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131253796","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}
Yan Mendes, Regina M. M. Braga, Victor Ströele, Daniela Alarcão De Oliveira
In the last decade the (big) data-driven science paradigm became a wide-spread reality. However, this approach has some limitations such as a performance dependency on the quality of the data and the lack of reproducibility of the results. In order to enable this reproducibility, many tools such as Workflow Management Systems were developed to formalize process pipelines and capture execution traces. However, interoperating data generated by these solutions became a problem, since most systems adopted proprietary data models. To support interoperability across heterogeneous provenance data, we propose a Service Oriented Architecture with a polystore storage design in which provenance is conceptually represented utilizing the ProvONE model. A wrapper layer is responsible for transforming data described by heterogeneous formats into ProvONE-compliant. Moreover, we propose a query layer that provides location and access transparency to users. Furthermore, we conduct two feasibility studies, showcasing real usecase scenarios. Firstly, we illustrate how two research groups can compare their processes and results. Secondly, we show how our architecture can be used as a queriable provenance repository. We show Polyflow's viability for both scenarios using the Goal-Question-Metric methodology. Finally, we show our solution usability and extensibility appeal by comparing it to similar approaches.
{"title":"Polyflow","authors":"Yan Mendes, Regina M. M. Braga, Victor Ströele, Daniela Alarcão De Oliveira","doi":"10.1145/3330204.3330259","DOIUrl":"https://doi.org/10.1145/3330204.3330259","url":null,"abstract":"In the last decade the (big) data-driven science paradigm became a wide-spread reality. However, this approach has some limitations such as a performance dependency on the quality of the data and the lack of reproducibility of the results. In order to enable this reproducibility, many tools such as Workflow Management Systems were developed to formalize process pipelines and capture execution traces. However, interoperating data generated by these solutions became a problem, since most systems adopted proprietary data models. To support interoperability across heterogeneous provenance data, we propose a Service Oriented Architecture with a polystore storage design in which provenance is conceptually represented utilizing the ProvONE model. A wrapper layer is responsible for transforming data described by heterogeneous formats into ProvONE-compliant. Moreover, we propose a query layer that provides location and access transparency to users. Furthermore, we conduct two feasibility studies, showcasing real usecase scenarios. Firstly, we illustrate how two research groups can compare their processes and results. Secondly, we show how our architecture can be used as a queriable provenance repository. We show Polyflow's viability for both scenarios using the Goal-Question-Metric methodology. Finally, we show our solution usability and extensibility appeal by comparing it to similar approaches.","PeriodicalId":348938,"journal":{"name":"Proceedings of the XV Brazilian Symposium on Information Systems","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115433958","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}
João Luiz Cavalcante Ferreira, André Filipe Aloise, V. Matter, Jorge L. V. Barbosa, S. Rigo, K. Oliveira
This paper proposes the MD-PREAD, a model that uses the decision tree technique for predicting apprentices with risk of failure. The capability of choosing the decision tree as a way to generate a greater set for educators is the highlight of this project. After the data was collected and processed, it was possible to generate a list of students that had the greatest chance to fail, this data would give the opportunity to help the students to recover their grades before the end of the course. Finally, to evaluate the model, the indexes of the classifiers were compared and the J48 algorithm stood out with an accuracy predominance of 84.5%, precision of 85.52%. It was concluded that the MD-PREAD model can assist in the prognosis of groups at risk of failure.
{"title":"A Model for Predicting Disapproval of Apprentices in Distance Education Using Decision Tree","authors":"João Luiz Cavalcante Ferreira, André Filipe Aloise, V. Matter, Jorge L. V. Barbosa, S. Rigo, K. Oliveira","doi":"10.1145/3330204.3330223","DOIUrl":"https://doi.org/10.1145/3330204.3330223","url":null,"abstract":"This paper proposes the MD-PREAD, a model that uses the decision tree technique for predicting apprentices with risk of failure. The capability of choosing the decision tree as a way to generate a greater set for educators is the highlight of this project. After the data was collected and processed, it was possible to generate a list of students that had the greatest chance to fail, this data would give the opportunity to help the students to recover their grades before the end of the course. Finally, to evaluate the model, the indexes of the classifiers were compared and the J48 algorithm stood out with an accuracy predominance of 84.5%, precision of 85.52%. It was concluded that the MD-PREAD model can assist in the prognosis of groups at risk of failure.","PeriodicalId":348938,"journal":{"name":"Proceedings of the XV Brazilian Symposium on Information Systems","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123446171","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：