摘要 :
The Paper-to-Reviewer Assignment Problem is a prevalent and challenging problem within the scientific community and scientific conferences. The large number of reviewers and submitted papers, as well as constraints on reviewers' l...
展开
The Paper-to-Reviewer Assignment Problem is a prevalent and challenging problem within the scientific community and scientific conferences. The large number of reviewers and submitted papers, as well as constraints on reviewers' loads and papers' coverage needs, make difficult and very tedious for the committee chair to manually assign submitted papers to suitable reviewers. Any automation of this task must tackle three main issues: modeling reviewers' and papers' profiles, estimating the relevance between a paper and a reviewer and finally finding an appropriate assignment.In this paper, we are interested in the two first issues. We propose to build the reviewers profiles on the basis of both pedagogical and research documents, then a formula that takes into account both topics similarity and common references is proposed. Experimental results compared to ground-truth values extracted from ICA2IT'19 data show a good performance of this approach.
收起
摘要 :
The Paper-to-Reviewer Assignment Problem is a prevalent and challenging problem within the scientific community and scientific conferences. The large number of reviewers and submitted papers, as well as constraints on reviewers' l...
展开
The Paper-to-Reviewer Assignment Problem is a prevalent and challenging problem within the scientific community and scientific conferences. The large number of reviewers and submitted papers, as well as constraints on reviewers' loads and papers' coverage needs, make difficult and very tedious for the committee chair to manually assign submitted papers to suitable reviewers. Any automation of this task must tackle three main issues: modeling reviewers' and papers' profiles, estimating the relevance between a paper and a reviewer and finally finding an appropriate assignment.In this paper, we are interested in the two first issues. We propose to build the reviewers profiles on the basis of both pedagogical and research documents, then a formula that takes into account both topics similarity and common references is proposed. Experimental results compared to ground-truth values extracted from ICA2IT'19 data show a good performance of this approach.
收起
摘要 :
Opportunities to conduct the complicated documenting work, that is, creating and elaborating documents through such tasks as exploring the Web and reviewing existing documents, have increased more and more. For such work, it is im...
展开
Opportunities to conduct the complicated documenting work, that is, creating and elaborating documents through such tasks as exploring the Web and reviewing existing documents, have increased more and more. For such work, it is important to skillfully manage the deliverables along with their editing processes and utilize them for later work. However, the documents involved generally become diversified and enormous in number since documenting work is generally done over a long period of time. These features make managing documents difficult. In particular, it is often the case that a user wants to find and reuse documents that he/she used in past work. It is however difficult to recall past work situations and the documents involved on the basis of human memory and folder management. In this research, we aim at developing methods for organizing work situations and the documents involved via intuitive clues. In this paper, we describe a framework that supports complicated documenting work. We also describe methods for generating recall networks. Finally, we describe an experiment done using practical history data and discuss the characteristics of our methods based on the results.
收起
摘要 :
Opportunities to conduct the complicated documenting work, that is, creating and elaborating documents through such tasks as exploring the Web and reviewing existing documents, have increased more and more. For such work, it is im...
展开
Opportunities to conduct the complicated documenting work, that is, creating and elaborating documents through such tasks as exploring the Web and reviewing existing documents, have increased more and more. For such work, it is important to skillfully manage the deliverables along with their editing processes and utilize them for later work. However, the documents involved generally become diversified and enormous in number since documenting work is generally done over a long period of time. These features make managing documents difficult. In particular, it is often the case that a user wants to find and reuse documents that he/she used in past work. It is however difficult to recall past work situations and the documents involved on the basis of human memory and folder management. In this research, we aim at developing methods for organizing work situations and the documents involved via intuitive clues. In this paper, we describe a framework that supports complicated documenting work. We also describe methods for generating recall networks. Finally, we describe an experiment done using practical history data and discuss the characteristics of our methods based on the results.
收起
摘要 :
In this paper a new probabilistic topic model is introduced which can provides two levels of topics called specific and general topics. LDA model as a basic probabilistic topic model suffers from over-generalization and to the bes...
展开
In this paper a new probabilistic topic model is introduced which can provides two levels of topics called specific and general topics. LDA model as a basic probabilistic topic model suffers from over-generalization and to the best of our knowledge this problem is unsolved. By adding another level of topics we tried to overcome this problem. The proposed model is applied to a corpus of 250 documents and 6143 unique words. The results show that the proposed model can produce more specific topics and also can produce clusters that are more similar to human-assigned categories.
收起
摘要 :
In this paper a new probabilistic topic model is introduced which can provides two levels of topics called specific and general topics. LDA model as a basic probabilistic topic model suffers from over-generalization and to the bes...
展开
In this paper a new probabilistic topic model is introduced which can provides two levels of topics called specific and general topics. LDA model as a basic probabilistic topic model suffers from over-generalization and to the best of our knowledge this problem is unsolved. By adding another level of topics we tried to overcome this problem. The proposed model is applied to a corpus of 250 documents and 6143 unique words. The results show that the proposed model can produce more specific topics and also can produce clusters that are more similar to human-assigned categories.
收起
摘要 :
Electronic Document and Records Management System (EDRMS) is recognized to have a positive impact on the management of organization records. However, the system has not been optimally utilized because of rejection among the users....
展开
Electronic Document and Records Management System (EDRMS) is recognized to have a positive impact on the management of organization records. However, the system has not been optimally utilized because of rejection among the users. The adoption of the system is influenced by ten identified factors that are then used as a basis in questionnaire development. In this regard, the study aims to develop an authoritative questionnaire following stringent instrument development protocols. There were 72 indicators identified. After content validity measurement performed, seven indicators were rejected, and only 65 indicators were accepted to form the final questionnaire. The validated and evaluated questionnaire can be used to measure the level of EDRMS adoption in an organization.
收起
摘要 :
Electronic Document and Records Management System (EDRMS) is recognized to have a positive impact on the management of organization records. However, the system has not been optimally utilized because of rejection among the users....
展开
Electronic Document and Records Management System (EDRMS) is recognized to have a positive impact on the management of organization records. However, the system has not been optimally utilized because of rejection among the users. The adoption of the system is influenced by ten identified factors that are then used as a basis in questionnaire development. In this regard, the study aims to develop an authoritative questionnaire following stringent instrument development protocols. There were 72 indicators identified. After content validity measurement performed, seven indicators were rejected, and only 65 indicators were accepted to form the final questionnaire. The validated and evaluated questionnaire can be used to measure the level of EDRMS adoption in an organization.
收起
摘要 :
This paper presents an overview of selected clustering models and shows an application of K-Means algorithm to document clustering. In the introductory part, the definitions of basic concepts and common characteristics of clusteri...
展开
This paper presents an overview of selected clustering models and shows an application of K-Means algorithm to document clustering. In the introductory part, the definitions of basic concepts and common characteristics of clustering models are described. Then an overview of clustering models is given. The methods of clustering, basic characteristics, visualization and possible input data for each algorithm are presented. The authors also explain the assessment of each algorithm taking into consideration measures such as Rand index, homogeneity completeness, V-measure and Silhouette coefficient. Furthermore, the paper describes the application of the K-Means algorithm to document clustering showing the final result and elaborating the procedures applied when clustering the documents.
收起
摘要 :
This paper presents an overview of selected clustering models and shows an application of K-Means algorithm to document clustering. In the introductory part, the definitions of basic concepts and common characteristics of clusteri...
展开
This paper presents an overview of selected clustering models and shows an application of K-Means algorithm to document clustering. In the introductory part, the definitions of basic concepts and common characteristics of clustering models are described. Then an overview of clustering models is given. The methods of clustering, basic characteristics, visualization and possible input data for each algorithm are presented. The authors also explain the assessment of each algorithm taking into consideration measures such as Rand index, homogeneity completeness, V-measure and Silhouette coefficient. Furthermore, the paper describes the application of the K-Means algorithm to document clustering showing the final result and elaborating the procedures applied when clustering the documents.
收起