Montse Cuadros

Quality Assessment of Large Scale Knowledge Resources

This paper presents an empirical evaluation of the quality of publicly available large-scale knowledge resources. The study includes a wide range of manually and automatically derived large-scale knowledge resources. In order to establish a fair and neutral comparison, the quality of each knowledge resource is indirectly evaluated using the same method on a Word Sense Disambiguation task. The evaluation framework selected has been the Senseval-3 English Lexical Sample Task. The study empirically demonstrates that automatically acquired knowledge resources surpass both in terms of precision and recall the knowledge resources derived manually, and that the combination of the knowledge contained in these resources is very close to the most frequent sense classifier. As far as we know, this is the first time that such a quality assessment has been performed showing a clear picture of the current state-of-the-art of publicly available wide coverage semantic resources.

KnowNet: Building a Large Net of Knowledge from the Web

This paper presents a new fully automatic method for building highly dense and accurate knowledge bases from existing semantic resources. Basically, the method uses a wide-coverage and accurate knowledge-based Word Sense Disambiguation algorithm to assign the most appropriate senses to large sets of topically related words acquired from the web. KnowNet, the resulting knowledge-base which connects large sets of semantically-related concepts is a major step towards the autonomous acquisition of knowledge from raw corpora. In fact, KnowNet is several times larger than any available knowledge resource encoding relations between synsets, and the knowledge KnowNet contains outperform any other resource when is empirically evaluated in a common framework.

W2VLDA: Almost unsupervised system for Aspect Based Sentiment Analysis

With the increase of online customer opinions in specialised websites and social networks, the necessity of automatic systems to help to organise and classify customer reviews by domain-specific aspect/categories and sentiment polarity is more important than ever. Supervised approaches to Aspect Based Sentiment Analysis obtain good results for the domain/language their are trained on, but having manually labelled data for training supervised systems for all domains and languages are usually very costly and time consuming. In this work we describe W2VLDA, an almost unsupervised system based on topic modelling, that combined with some other unsupervised methods and a minimal configuration, performs aspect/category classifiation, aspect-terms/opinion-words separation and sentiment polarity classification for any given domain and language. We evaluate the performance of the aspect and sentiment classification in the multilingual SemEval 2016 task 5 (ABSA) dataset. We show competitive results for several languages (English, Spanish, French and Dutch) and domains (hotels, restaurants, electronic-devices).

Automatic analysis of textual hotel reviews

Social Media and consumer-generated content continue to grow and impact the hospitality domain. Consumers write online reviews to indicate their level of satisfaction with a hotel and inform other consumers on the Internet of their hotel stay experience. A number of websites specialized in tourism and hospitality have flourished on the Web (e.g. Tripadvisor). The tremendous growth of these data-generating sources demands new tools to deal with them. To cope with big amounts of customer-generated reviews and comments, Natural Language Processing (NLP) tools have become necessary to automatically process and manage textual customer reviews (e.g. to perform Sentiment Analysis). This work describes OpeNER, a NLP platform applied to the hospitality domain to automatically process customer-generated textual content and obtain valuable information from it. The presented platform consists of a set of Open Source and free NLP tools to analyse text based on a modular architecture to ease its modification and extension. The training and evaluation has been performed using a set of manually annotated hotel reviews gathered from websites like Zoover and HolidayCheck.

SemEval-2007 Task 16: Evaluation of Wide Coverage Knowledge Resources

This task tries to establish the relative quality of available semantic resources (derived by manual or automatic means). The quality of each large-scale knowledge resource is indirectly evaluated on a Word Sense Disambiguation task. In particular, we use Senseval-3 and SemEval-2007 English Lexical Sample tasks as evaluation bechmarks to evaluate the relative quality of each resource. Furthermore, trying to be as neutral as possible with respect the knowledge bases studied, we apply systematically the same disambiguation method to all the resources. A completely different behaviour is observed on both lexical data sets (Senseval-3 and SemEval-2007).

V3: Unsupervised Aspect Based Sentiment Analysis for SemEval2015 Task 12

This paper presents our participation in SemEval-2015 task 12 (Aspect Based Sentiment Analysis). We participated employing only unsupervised or weakly-supervised approaches. Our attempt is based on requiring the minimum annotated or hand-crafted content, and avoids training a model using the provided training set. We use a continuous word representations (Word2Vec) to leverage in-domain semantic similarities of words for many of the involved subtasks.

NERC-fr: Supervised Named Entity Recognition for French

Currently there are only few available language resources for French. Additionally there is a lack of available language models for for tasks such as Named Entity Recognition and Classification (NERC) which makes difficult building natural language processing systems for this language. This paper presents a new publicly available supervised Apache OpenNLP NERC model that has been trained and tested under a maximum entropy approach. This new model achieves state of the art results for French when compared with another systems. Finally we have also extended Apache OpenNLP libraries to support part-of-speech feature extraction component which has been used for our experiments.

V3: Unsupervised Generation of Domain Aspect Terms for Aspect Based Sentiment Analysis

This paper presents V3, an unsupervised system for aspect-based Sentiment Analysis when evaluated on the SemEval 2014 Task 4. V3 focuses on generating a list of aspect terms for a new domain using a collection of raw texts from the domain. We also implement a very basic approach to classify the aspect terms into categories and assign polarities to them.

Correlating Languages and Sentiment Analysis on the Basis of Text-based Reviews

Customer experiences, in the shape of online reviews, influence other customers and in general, contribute to build a perception of a destination. This work presents the conclusions of a survey to gather user text-based reviews about several categories of destination-related information (accommodation, restaurants, attractions and Points of Interest) from three well-known social media sources (Facebook, FourSquare and GooglePlaces) about eight worldwide destinations with a high overnight rate. Several hypotheses about the correlation between the language and sentiment features of the reviews have been validated over a large dataset of reviews. For example, the analysis detected that the highest number of reviews in a destination is written in the same official language spoken in that place. Furthermore, Dutch speaking people are more positive when writing a review. Finally, English, Italian and Spanish speakers seem to prefer FourSquare while German and French people are quite evenly distributed among FourSquare and GooglePlaces.

KnowNet: A Proposal for Building Highly Connected and Dense Knowledge Bases from the Web

This paper presents a new fully automatic method for building highly dense and accurate knowledge bases from existing semantic resources. Basically, the method uses a wide-coverage and accurate knowledge-based Word Sense Disambiguation algorithm to assign the most appropriate senses to large sets of topically related words acquired from the web. KnowNet, the resulting knowledge-base which connects large sets of semantically-related concepts is a major step towards the autonomous acquisition of knowledge from raw corpora. In fact, KnowNet is several times larger than any available knowledge resource encoding relations between synsets, and the knowledge that KnowNet contains outperform any other resource when empirically evaluated in a common multilingual framework.