Fernando Martínez-Santiago

Crowd explicit sentiment analysis

With the rapid growth of data generated by social web applications new paradigms in the generation of knowledge are opening. This paper introduces Crowd Explicit Sentiment Analysis (CESA) as an approach for sentiment analysis in social media environments. Similar to Explicit Semantic Analysis, microblog posts are indexed by a predefined collection of documents. In CESA, these documents are built up from common emotional expressions in social streams. In this way, texts are projected to feelings or emotions. This process is performed within a Latent Semantic Analysis. A few simple regular expressions (e.g. “I feel X”, considering X a term representing an emotion or feeling) are used to scratch the enormous flow of micro-blog posts to generate a textual representation of an emotional state with clear polarity value (e.g. angry, happy, sad, confident, etc.). In this way, new posts can be indexed by these feelings according to the distance to their textual representation. The approach is suitable in many scenarios dealing with social media publications and can be implemented in other languages with little effort. In particular, we have evaluated the system on Polarity Classification with both English and Spanish data sets. The results show that CESA is a valid solution for sentiment analysis and that similar approaches for model building from the continuous flow of posts could be exploited in other scenarios.

A merging strategy proposal: The 2-step retrieval status value method

A usual strategy to implement CLIR (Cross-Language Information Retrieval) systems is the so-called query translation approach. The user query is translated for each language present in the multilingual collection in order to compute an independent monolingual information retrieval process per language. Thus, this approach divides documents according to language. In this way, we obtain as many different collections as languages. After searching in these corpora and obtaining a result list per language, we must merge them in order to provide a single list of retrieved articles.In this paper, we propose an approach to obtain a single list of relevant documents for CLIR systems driven by query translation. This approach, which we call 2-step RSV (RSV: Retrieval Status Value), is based on the re-indexing of the retrieval documents according to the query vocabulary, and it performs noticeably better than traditional methods.The proposed method requires query vocabulary alignment: given a word for a given query, we must know the translation or translations to the other languages. Because this is not always possible, we have researched on a mixed model. This mixed model is applied in order to deal with queries with partial word-level alignment. The results prove that even in this scenario, 2-step RSV performs better than traditional merging methods.

Otium: A web based planner for tourism and leisure

This paper introduces the Otiŭm planner system for scheduling of leisure tasks in tourism. This novel service allows users to create their own agenda of activities within specified dates. Activities are selected from a list of recommended events according to last selected events, user preferences and other parameters. The proposed restrictions on the recommendation procedure have been found to capture static and dynamic user context. The recommendation function is linear and shows low computational cost. The events are extracted from web sources with almost no human manipulation, so the recommender is always showing new and recent events. The Ajax-based web interface eases the creation of the final plan, offering an interactive experience to the user. We consider that the trade-off between interactivity and recommendation complexity exits, and that the second issue is preferable in this type of services. The details about the design and implementation of the system are described, along with the issues the system resolves and some guidelines for enhancement. 2011 Elsevier Ltd. All rights reserved.

SINAI at CLEF Ad-Hoc Robust Track 2007: Applying Google Search Engine for Robust Cross-Lingual Retrieval

We report our web-based query generation experiments for English and French collections in the Robust task of the CLEF Ad-Hoc track. We continued with the approach adopted in the previous year, although the model has been modified. Last year we used Google to expand the original query. This year we create a new expanded query in addition to the original one. Thus, we retrieve two lists of relevant documents, one for each query (the original and the expanded one). In order to integrate the two lists of documents, we apply a logistic regression merging solution. The results obtained are discouraging but the failure analysis shows that very difficult queries are improved by using both queries instead of the original query. The problem is to decide when a query is very difficult.We report our web-based query generation experiments for English and French collections in the Robust task of the CLEF Ad-Hoc track. We continued with the approach adopted in the previous year, although the model has been modified. Last year we used Google to expand the original query. This year we create a new expanded query in addition to the original one. Thus, we retrieve two lists of relevant documents, one for each query (the original and the expanded one). In order to integrate the two lists of documents, we apply a logistic regression merging solution. The results obtained are discouraging but the failure analysis shows that very difficult queries are improved by using both queries instead of the original query. The problem is to decide when a query is very difficult.

SINAI at CLEF 2003: decompounding and merging

This paper describes the application of the two-step RSV and mixed two-step RSV merging methods in the multilingual-4 and multilingual-8 tasks at CLEF 2003. We study the performance of these methods compared to previous studies and approaches. A new strategy for dealing with compound words which uses predefined vocabularies for automatic decomposition is also presented and evaluated.

Evaluating word sense disambiguation tools for information retrieval task

The main interest of this paper is the characterization of queries where WSD is a useful tool. That is, which issues must be fulfilled by a query in order to apply an state-of-art WSD tool? In addition, we have evaluated several approaches in order to apply WSD. We have used several types of indices. Thus, we have generated 13 indices and we have carried out 39 different experiments, obtaining that some indices based on WSD tools even outperforms slightly the non disambiguated baseline case. After the interpretation of our experiments, we think that only queries with terms very polysemous and very high IDF value are improved by using WSD.

Combining Lexical-Syntactic Information with Machine Learning for Recognizing Textual Entailment

This document contains the description of the experiments carried out by SINAI group. We have developed an approach based on several lexical and syntactic measures integrated by means of different machine learning models. More precisely, we have evaluated three features based on lexical similarity and 11 features based on syntactic tree comparison. In spite of the relatively straightforward approach we have obtained more than 60% for accuracy. Since this is our first participation we think we have reached a good result.

SINAI at CLEF 2006 ad hoc robust multilingual track: query expansion using the Google search engine

This year, we have participated in the Ad-Hoc Robust Multilingual track with the aim of evaluating two important issues in Cross-Lingual Information Retrieval (CLIR) systems. This paper first describes the method applied for query expansion in a multilingual environment by using web search results provided by the Google engine in order to increase retrieval robustness. Unfortunately, the results obtained are disappointing. The second issue reported alludes to the robustness of several common merging algorithms. We have found that 2-step RSV merging algorithms perform better than others algorithms when evaluating using geometric average.This year, we have participated on Ad-Hoc Robust Multilingual track with the aim to evaluate two issues of CLIR systems. Firstly, this paper describes the method followed for query expansion in a multilingual environment by using web search results provided by the Google engine in order to increment retrieval robustness. Unfortunately, the results obtained are disappointing. The second issue reported is relative to the robustness of several usual merging algorithms. We have found that 2-step RSV merging algorithms perform better than others algorithms when geometric precision is applied.

SINAI at CLEF 2004: using machine translation resources with a mixed 2-step RSV merging algorithm

In CLEF 2004, the SINAI group participated in the multilingual task. Our main interest was to test Machine Translation (MT) with a mixed 2-step RSV merging algorithm. Since 2-step RSV requires grouping the document frequency for each term with the translations for that term, and MT translates whole phrases better than working word for word, it is not directly feasible to use MT with a 2-step RSV merging algorithm. To solve this problem, we have tested an algorithm which aligns the original query and its translation(s) at term level.

Merging Strategy for Cross-Lingual Information Retrieval Systems based on Learning Vector Quantization

We present a new approach based on neural networks to solve the merging strategy problem for Cross-Lingual Information Retrieval (CLIR). In addition to language barrier issues in CLIR systems, how to merge a ranked list that contains documents in different languages from several text collections is also critical. We propose a merging strategy based on competitive learning to obtain a single ranking of documents merging the individual lists from the separate retrieved documents. The main contribution of the paper is to show the effectiveness of the Learning Vector Quantization (LVQ) algorithm in solving the merging problem. In order to investigate the effects of varying the number of codebook vectors, we have carried out several experiments with different values for this parameter. The results demonstrate that the LVQ algorithm is a good alternative merging strategy.