Abstract

Community Question Answering (cQA) forums have become a popular medium for soliciting answers to specific user questions from experts and experienced users in a given topic. However,for a given question, users sometimes have to sift through a large number of low-quality or irrelevant answers to find out the answer which satisfies their information need. To alleviate this,the problem of Answer Quality Prediction (AQP) aims to predict the quality of an answer posted in response to a forum question. Current AQP systems either learn models using - a) various hand-crafted features (HCF) or b) Deep Learning (DL) techniques which automatically learn the feature representations.In this paper, we propose a novel approach for AQP known as -“Deep Feature Fusion Network(DFFN)”which combines the advantages of both hand-crafted features and deep learning based systems. Given a question-answer pair along with its metadata, a DFFN architecture indepen-dently - a) learns features using the Deep Neural Network (DNN) and b) computes handcrafted features leveraging various external resources and then combines the musing a fully connected neural network trained to predict the quality of the given answer. DFFN is an end-end differen-tiable model and trained as a single system. We propose two different DFFN architectures which vary mainly in the way they model the input question/answer pair - a) DFFN-CNN which uses a Convolutional Neural Network (CNN) and b) DFFN-BLNA which uses a Bi-directional LSTMwith Neural Attention (BLNA). Both these proposed variants of DFFN (DFFN-CNN and DFFN-BLNA) achieve state-of-the-art performance on the standard SemEval-2015 and SemEval-2016benchmark datasets and outperforms baseline approaches which individually employ either HCFor DL based techniques alone.

Abstract

In this paper, we introduce Vaidya, a spoken dialog system which is developed as part of the ITRA1project. The system is capable of providing an approximate diagnosis by accepting symptoms as free-form speech in real-time on both laptop and hand-held devices. The system focuses on challenges in speech recognition specific to Indian languages and capturing the intent of the user. Another challenge is to create models which are memory and CPU efficient for hand-held devices. We describe our progress, experiences and approaches in building the system that can handle English as the input speech. The system is evaluated using subjective statistical measure (Fleiss’ kappa) to assess the usability of the system

Abstract

We present Kathaa, an Open Source web-based Visual Programming Framework for Natural Language Processing (NLP) Systems. Kathaa supports the design, execution and analysis of complex NLP systems by visually connecting NLP components from an easily extensible Module Library. It models NLP systems an edge-labeled Directed Acyclic MultiGraph, and lets the user use publicly co-created modules in their own NLP applications irrespective of their technical proficiency in Natural Language Processing. Kathaa exposes an intuitive web based Interface forthe users to interact with and modify complex NLP Systems; and a precise Module definition API to allow easy integration of new state of the art NLP components. Kathaa enables researchers to publish their services in a standardized format to enable the masses to use their services out ofthe box. The vision of this work is to pave the way for a system like Kathaa, to be the Lego blocks of NLP Research and Applications. As a practical use case we use Kathaa to visually implement the Sampark Hindi-Panjabi Machine Translation Pipeline and the Sampark Hindi-Urdu Machine Translation Pipeline, to demonstrate the fact that Kathaa can handle really complex NLP systems while still being intuitive for the end user.

Abstract

Designing various learning strategies has been gaining a lot of scientific interest during the recent progress of deep learning methodologies. Curriculum learning is a learning strategy aimed at training the neural network model by presenting the samples in a specific meaningful order rather than randomly sampling the training examples from the data distribution. In this work, we have explored starting small paradigm of curriculum learning technique for speech recognition. The starting small paradigm of curriculum learning is performed by a two step learning strategy. Training dataset is re-organized as a set of easily classifiable examples followed by the actual training dataset and the model is trained on the re-organized dataset. We hypothesize that by following the starting small learning paradigm the learning gets initialized in a better way and progresses to attain a better convergence. We propose to rank the toughness of the training example based on the posterior probabilities obtained using a previously trained model. Apart from re-arranging the training corpus starting small paradigm of curriculum learning is applied at model level. We consider the broad manner-class classification objective function as the smoother version of the phone class classification objective function. The model initially trained for broad class classification is later adapted for phone classification.In this work, we have used TIMIT and a subset of WallStreet Journal (WSJ) corpus to validate the experiments, both the learning strategies have shown consistently better performances across the two datasets compared to the baseline system trained by randomly sampling the dataset.

Abstract

n this study, the problem of shallow parsing of Hindi-English code-mixed social media text (CSMT) has been addressed. We have annotated the data, developed a language identifier, a normalizer, a part-of-speech tagger and a shallow parser. To the best of our knowledge, we are the first to attempt shallow parsing on CSMT. The pipeline developed has been made available to the research community with the goal of enabling better text analysis of Hindi English CSMT. The pipeline is accessible at this http URL.

Abstract

Sentiment analysis (SA) using code-mixed data from social media has several applications in opinion mining ranging from customer satisfaction to social campaign analysis in multilingual societies. Advances in this area are impeded by the lack of a suitable annotated dataset. We introduce a Hindi-English (Hi-En) code-mixed dataset for sentiment analysis and perform empirical analysis comparing the suitability and performance of various state-of-the-art SA methods in social media. In this paper, we introduce learning sub-word level representations in our LSTM (Subword-LSTM) architecture instead of character-level or word-level representations. This linguistic prior in our architecture enables us to learn the information about sentiment value of important morphemes. This also seems to work well in highly noisy text containing misspellings as shown in our experiments which is demonstrated in morpheme-level feature maps learned by our model. Also, we hypothesize that encoding this linguistic prior in the Subword-LSTM architecture leads to the superior performance. Our system attains accuracy 4-5% greater than traditional approaches on our dataset, and also outperforms the available system for sentiment analysis in Hi-En code-mixed text by 18%.

Abstract

Automating the process of indexing journal abstracts has been a topic of research for several years. Biomedical Semantic Indexing aims to assign correct MeSH terms to the PubMed documents. In this paper we report our participation in the Task 3a of BioASQ challenge 2015. The participating teams were provided with PubMed articles and asked to return relevant MeSH terms. We tried three different approaches: Nearest Neighbours, IDF-Ratio based indexing and multi-label classification. The official challenge results demonstrate that we consistently performed better than the baseline approaches for Task 3a.

Abstract

In this paper, we describe our participation in the 2015 BioASQ challenge on Bio-Medical Question Answering. For Question Answering task (Task 3b), teams were provided with natural language questions and asked to retrieve responses from PubMed corpus in the form of documents, snippets, concepts and RDF triplets (Phase A) and direct answers (Phase B). For Phase A, we took the support of PubMed search engine and our snippet extraction technique. In our QA system, apart from the standard techniques discussed in literature, we tried the following novel techniques to - a) leverage web search results for improving question processing and b) identify domain words and define a new answer ranking function based on number of common domain words. We scored an F-measure of 0.193 for document extraction and F-measure of 0.0717 in snippet generation

Abstract

We present a system employing domain specific dictionaries and features to recognize chemical entities. The system utilizes sentence segmentation, tokenization, feature generation, Conditional Random Field (CRF) training and one post-processing step. The dictio-naries were compiled from PubChem, Wikipedia, ChEMBL, DrugBank,word2vec clusters from US patents belonging to A61K class. We report the evaluation results of the run where development set was not included as part of the training set. The best performing model for CEMP taskh as the micro average precision, recall and F-score values of87.26%,79.98%and83.46%, respectively.

Abstract

Sindhi is an Indo-Aryan language spoken by more than 58 million speakers around the world. It is currently a resource poor language which is harmed by the literature being written in multiple scripts. Though the language is widely spoken,primarily, across two countries, the written form is not standardized. In this paper, we seek to develop resources for basic language processing for Sindhi language, in one of its preferred scripts (Devanagari), because a language that seeks to survive in the modern information society requires language technology products. This paper presents our work on building a stochastic Part-of-Speech tagger for Sindhi-Devanagari using conditional random fields with linguistically motivated features. The paper also discusses the steps taken to construct a part-of-speech annotated corpus for Sindhi in Devanagari script. We have also explained in detail the features that were used for training the tagger, which resulted in a part of speech tagger nearing 92% average accuracy.