A TV content augmentation system exploiting rule based named entity recognition method

---

Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Bilgisayar Mühendisliği Bölümü, Türkiye

Tezin Onay Tarihi: 2014

Öğrenci: YUNUS EMRE IŞIKLAR

Danışman: FEHİME NİHAN ÇİÇEKLİ

Özet:

In this thesis, a TV content augmentation system taking the advantage of named entity recognition methods is proposed. The system aims to automatically enhance TV program contents by retrieving context related data and presenting them to the viewers without any necessity of another device. In addition to conceptual description of the system, a prototype implementation is developed and demonstrated with predefined TV programs. The implementation utilizes Electronic Program Guide (EPG) data of programs crawled from web resources in order to extract named entities such as person names, locations, organizations, etc. For this purpose, a rule based Named Entity Recognition (NER) algorithm is developed for Turkish texts. Detailed information about the extracted entities is retrieved from Wikipedia after semantic disambiguation and its summarized form is presented to the users. A set of experiments have been conducted on two different data sets in order to evaluate the performance of the rule based NER algorithm and the behavior of the TV content augmentation system. The experimental results show that the content augmentation with NER methods is quite successful in TV domain especially for channels broadcasting news and series.