پديد آورنده :
زاهدي، اسماعيل
عنوان :
نظر كاوي با رويكرد مدل سازي ويژگي و احساس از داده هاي بزرگ زماني
مقطع تحصيلي :
كارشناسي ارشد
گرايش تحصيلي :
هوش مصنوعي
محل تحصيل :
اصفهان: دانشگاه صنعتي اصفهان، دانشكده برق و كامپيوتر
صفحه شمار :
دوازده، 121ص.: مصور
استاد راهنما :
محمدحسين سرايي
توصيفگر ها :
چارچوب كلان داده اسپارك , روش بدون نظارت
استاد داور :
محمدحسين منشئي، عبدالرضا ميرزايي
تاريخ ورود اطلاعات :
1395/07/10
دانشكده :
مهندسي برق و كامپيوتر
چكيده انگليسي :
Opinion Mining Toward joint Sentiment and Topic Modeling from Big Time Oreiented Data Esmaeil Zahedi e zahedi@ec iut ac ir Date of Submission 2016 04 30 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language FarsiSupervisor Dr Mohammad hossein Saraee saraee@cc iut ac ir Abstract In recent years people want to express their opinion on every online service or product and there arenow a huge number of opinions on the social media online stores and blogs However most of the opinionsare presented in plain text and thus require a powerful method to analyse this volume of unlabeled reviewsto obtain information about relevant details in minimum time and with a high accuracy In this thesis wepropose a supervised model to analyze large unlabeled opinion data sets This model has two phases preprocessing and a Supervised Sentiment and Aspect Model SSAM In the preprocessing phase we inputthousands of unlabeled opinions and received a set of key value pairs in which a key holds a word or anopinion and a value holds supervised information such as a sentiment label of this word or opinion Afterthat we give these pairs to the proposed SSAM algorithm which incorporates different levels of supervisedinformation such as document and sentence levels or document and term levels of supervised information to extract and cluster aspects related to a sentiment label and also classify opinions based on theirsentiments SSAM is implemented in a Spark big data framework to adapt to the explosive growth of onlineopinions We applied SSAM to reviews of electronic devices and books from Amazon The experimentsshow that the aspects found by proposed model capture important aspects that are closely coupled with asentiment label and also in sentiment classification proposed model outperforms other topic models andcomes close to supervised methods Keywords Big unlabeled opinion dataset Supervised Sentiment and Aspect Model Supervised andUnsupervised methods Supervised information
استاد راهنما :
محمدحسين سرايي
استاد داور :
محمدحسين منشئي، عبدالرضا ميرزايي
