MULTI-LINGUAL INFORMATION RETRIEVAL USING DEEP LEARNING

The task of finding data files related to an information need from a group of information resources is known as Information Retrieval. In this work, the author propose a multi-lingual information retrieval system using deep learning. Input to the system is a question in sentencing form that can be processed by NLP tools. In the preprocessing phase, part-of-speech tagging of the input sentence is performed. A three layer neural network is used for creating word to vector representation. The word2vec model continuous-bag-of-words (CBOW) is used for this purpose. Then related words are obtained via word-2-vec using deep learning RNN. RNN is the recurrent neural network. Finally, results are obtained by calculating the cosine similarity score. For multi-lingual results, bilingual mapping is performed using CFILTs bilingual corpus. The tourism dataset is used for experimentation purposes.