期刊名称：Indian Journal of Innovations and Developments
印刷版ISSN：2277-5382
电子版ISSN：2277-5390
出版年度：2016
卷号：5
期号：1
页码：1-10
语种：English
出版社：Indian Society for Education and Environment
摘要：Background/Objectives: To automatically extract policy networks by using Snippets and Social networks. Methods/Statistical analysis: The analysis of policy networks demands a series of difficult and time-consuming manual steps including interviews and questionnaires. The approach involves the process of estimating the strength of relations between actors in policy networks using features like webpage counts, out links, and lexical information extracted from data harvested from the web snippets. The approach extracts the irrelevant documents that affect the performance and accuracy. It is overcome by including the process of investigating machine learning algorithms for selecting the most informative metrics. Findings: The proposed approach includes metrics such as recovery degree, in-link, broken link, anchor text type and kl-divergence and filtering web data based on relevance and type of source, investigating the applicability of proposed metrics for social networks. This enhances the extraction of policy network more accurately. Improvements/Applications: Overall extraction of policy network is automatic and accurate while using the proposed approach of using Snippets and Social networks.
其他摘要：Background/Objectives: To automatically extract policy networks by using Snippets and Social networks. Methods/Statistical analysis: The analysis of policy networks demands a series of difficult and time-consuming manual steps including interviews and questionnaires. The approach involves the process of estimating the strength of relations between actors in policy networks using features like webpage counts, out links, and lexical information extracted from data harvested from the web snippets. The approach extracts the irrelevant documents that affect the performance and accuracy. It is overcome by including the process of investigating machine learning algorithms for selecting the most informative metrics. Findings: The proposed approach includes metrics such as recovery degree, in-link, broken link, anchor text type and kl-divergence and filtering web data based on relevance and type of source, investigating the applicability of proposed metrics for social networks. This enhances the extraction of policy network more accurately. Improvements/Applications: Overall extraction of policy network is automatic and accurate while using the proposed approach of using Snippets and Social networks.
关键词：Policy Networks; Relatedness Metrics; Spatial Proximity; Social Proximity.
