Web tabanlı yazılı iletişim karakteristikleri

In this thesis, we analyze different aspects of Web-based textual communicationsand argue that all such communications share some common properties. In order toprovide practical evidence for the validity of this argument, we focus on two common properties by examining these properties on various types of Web-based textualcommunications data. These properties are: All Web-based communications containfeatures attributable to their author and reciever; and all Web-based communicationsexhibit similar heavy tailed distributional properties.In order to provide practical proof for the validity of our claims, we provide threepractical, real life research problems and exploit the proposed common properties ofWeb-based textual communications to ?nd practical solutions to these problems. Inthis work, we ?rst provide a feature-based result caching framework for real life searchengines. To this end, we mined attributes from user queries in order to classify queriesand estimate a quality metric for giving admission and eviction decisions for the queryresult cache. Second, we analyzed messages of an online chat server in order to predictuser and mesage attributes. Our results show that several user- and message-basedattributes can be predicted with signi?cant occuracy using both chat message- andwriting-style based features of the chat users. Third, we provide a parallel frameworkfor in-memory construction of term partitioned inverted indexes. In this work, in orderto minimize the total communication time between processors, we provide a bucketingscheme that is based on term-based distributional properties of Web page contents.Keywords: Web search engine, result caching, cache, chat mining, data mining, indexinversion, inverted index, posting list.