عنوان سخنرانی: مدل هاي يادگيري عميق براي دسته بندي عناصر

سخنران: آقاي يداله يعقوب زاده، دانشجوي دكتري دانشگاه مونيخ

تاریخ برگزاری: 1395/12/23

فایل صوتی سخنرانی: 

Abstract: Recently, deep learning (DL) has been successful in many machine learning applications, including different natural language processing (NLP) problems. One of the important areas of NLP is information extraction (IE), which deals with structuring the text data into knowledge. This knowledge consists of various information about entities such as their properties, types, and their relations. In this talk, I will focus on the types (or classes) of entities and ways to extract them from text using DL models. I will introduce the problem of fine-grained entity typing, i.e., inferring from a large corpus of text that an entity is a member of a class such as “food” or “artist”. I will then present our two DL based approaches to tackle this problem: (i) a global model that types entities by learning aggregated representations (embeddings) of them (ii) a context model that scores each individual context (represented by word embeddings) of entities and then aggregates them.