{"title":"A Method to Identify Relevant Information Sufficient to Answer Situation Dependent Queries","authors":"Shan Lu, M. Kokar","doi":"10.1109/COGSIMA.2018.8423973","DOIUrl":null,"url":null,"abstract":"In various complex and dynamic environments, having a good understanding of the current situation in hand is the foundation for successful decision-making. Several frameworks have been proposed for information gathering and interpretation in situation assessment. However, decision makers nowadays face an information overload challenge during situation assessment. When the decision maker deals with a specific situation, usually large volumes of information are delivered to him or her in real time, of which only a few are relevant. It is practically impossible for them to deal with such huge data streams in real time. Additionally, if a situation needs to be communicated to others, it is not clear what information is relevant and thus would need to be sent over (sometimes over-loaded) communication links in order to convey the description of the situation. Therefore, a method is needed to support the human decision makers to identify the relevant information in situation assessment. In this paper, we develop a inference-based information relevance reasoning method in situation assessment to automatically identify relevant information for characterizing the situation that a decision maker is dealing with. By using this method, the following two basic questions will be answered: (1) what kind of information is relevant to characterize a situation? (2) how to identify the relevant information automatically? In this paper, we take the cyber security as the application domain, and evaluate our method using a cyber security dataset generated by Skaion corporation. We use four metrics to evaluate our method.","PeriodicalId":231353,"journal":{"name":"2018 IEEE Conference on Cognitive and Computational Aspects of Situation Management (CogSIMA)","volume":"127 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 IEEE Conference on Cognitive and Computational Aspects of Situation Management (CogSIMA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/COGSIMA.2018.8423973","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2
Abstract
In various complex and dynamic environments, having a good understanding of the current situation in hand is the foundation for successful decision-making. Several frameworks have been proposed for information gathering and interpretation in situation assessment. However, decision makers nowadays face an information overload challenge during situation assessment. When the decision maker deals with a specific situation, usually large volumes of information are delivered to him or her in real time, of which only a few are relevant. It is practically impossible for them to deal with such huge data streams in real time. Additionally, if a situation needs to be communicated to others, it is not clear what information is relevant and thus would need to be sent over (sometimes over-loaded) communication links in order to convey the description of the situation. Therefore, a method is needed to support the human decision makers to identify the relevant information in situation assessment. In this paper, we develop a inference-based information relevance reasoning method in situation assessment to automatically identify relevant information for characterizing the situation that a decision maker is dealing with. By using this method, the following two basic questions will be answered: (1) what kind of information is relevant to characterize a situation? (2) how to identify the relevant information automatically? In this paper, we take the cyber security as the application domain, and evaluate our method using a cyber security dataset generated by Skaion corporation. We use four metrics to evaluate our method.