Rocco Tripodi

A game-theoretic approach to word sense disambiguation

This article presents a new model for word sense disambiguation formulated in terms of evolutionary game theory, where each word to be disambiguated is represented as a node on a graph whose edges represent word relations and senses are represented as classes. The words simultaneously update their class membership preferences according to the senses that neighboring words are likely to choose. We use distributional information to weigh the influence that each word has on the decisions of the others and semantic similarity information to measure the strength of compatibility among the choices. With this information we can formulate the word sense disambiguation problem as a constraint satisfaction problem and solve it using tools derived from game theory, maintaining the textual coherence. The model is based on two ideas: Similar words should be assigned to similar classes and the meaning of a word does not depend on all the words in a text but just on some of them. The article provides an in-depth motivation of the idea of modeling the word sense disambiguation problem in terms of game theory, which is illustrated by an example. The conclusion presents an extensive analysis on the combination of similarity measures to use in the framework and a comparison with state-of-the-art systems. The results show that our model outperforms state-of-the-art algorithms and can be applied to different tasks and in different scenarios.

Document Clustering Games

In this article we propose a new model for document clustering, based on game theoretic principles. Each document to be clustered is represented as a player, in the game theoretic sense, and each cluster as a strategy that the players have to choose in order to maximize their payoff. The geometry of the data is modeled as a graph, which encodes the pairwise similarity among each document and the games are played among similar players. In each game the players update their strategies, according to what strategy has been effective in previous games. The Dominant Set clustering algorithm is used to find the prototypical elements of each cluster. This information is used in order to divide the players in two disjoint sets, one collecting labeled players, which always play a definite strategy and the other one collecting unlabeled players, which update their strategy at each iteration of the games. The evaluation of the system was conducted on 13 document datasets and shows that the proposed method performs well compared to different document clustering algorithms.

WSD-games: a Game-Theoretic Algorithm for Unsupervised Word Sense Disambiguation

In this paper we present an unsupervised approach to word sense disambiguation based on evolutionary game theory. In our algorithm each word to be disambiguated is represented as a node on a graph and each sense as a class. The algorithm performs a consistent class assignment of senses according to the similarity information of each word with the others, so that similar words are constrained to similar classes. The dynamics of the system are formulated in terms of a non-cooperative multiplayer game, where the players are the data points to decide their class memberships and equilibria correspond to consistent labeling of the data.

Document Clustering Games in Static and Dynamic Scenarios

In this work we propose a game theoretic model for document clustering. Each document to be clustered is represented as a player and each cluster as a strategy. The players receive a reward interacting with other players that they try to maximize choosing their best strategies. The geometry of the data is modeled with a weighted graph that encodes the pairwise similarity among documents, so that similar players are constrained to choose similar strategies, updating their strategy preferences at each iteration of the games. We used different approaches to find the prototypical elements of the clusters and with this information we divided the players into two disjoint sets, one collecting players with a definite strategy and the other one collecting players that try to learn from others the correct strategy to play. The latter set of players can be considered as new data points that have to be clustered according to previous information. This representation is useful in scenarios in which the data are streamed continuously. The evaluation of the system was conducted on 13 document datasets using different settings. It shows that the proposed method performs well compared to different document clustering algorithms.

Context aware nonnegative matrix factorization clustering

In this article we propose a method to refine the clustering results obtained with the nonnegative matrix factorization (NMF) technique, imposing consistency constraints on the final labeling of the data. The research community focused its effort on the initialization and on the optimization part of this method, without paying attention to the final cluster assignments. We propose a game theoretic framework in which each object to be clustered is represented as a player, which has to choose its cluster membership. The information obtained with NMF is used to initialize the strategy space of the players and a weighted graph is used to model the interactions among the players. These interactions allow the players to choose a cluster which is coherent with the clusters chosen by similar players, a property which is not guaranteed by NMF, since it produces a soft clustering of the data. The results on common benchmarks show that our model is able to improve the performances of many NMF formulations.

From Logical Forms to SPARQL Query with GETARUNS

We present a system for Question Answering which computes a prospective answer from Logical Forms produced by a full-fledged NLP for text understanding, and then maps the result onto schemata in SPARQL to be used for accessing the Semantic Web. As an intermediate step, and whenever there are complex concepts to be mapped, the system looks for a corresponding amalgam in YAGO classes. It is just by the internal structure of the Logical Form that we are able to produce a suitable and meaningful context for concept disambiguation. Logical Forms are the final output of a complex system for text understanding - GETARUNS - which can deal with different levels of syntactic and semantic ambiguity in the generation of a final structure, by accessing computational lexical equipped with sub-categorization frames and appropriate selectional restrictions applied to the attachment of complements and adjuncts. The system also produces pronominal binding and instantiates the implicit arguments, if needed, in order to complete the required Predicate Argument structure which is licensed by the semantic component.

Transductive Learning Games for Word Sense Disambiguation

Abstract: Word Sense Disambiguation (WSD) is the task of identifying the intended sense of a word in a computational manner based on the context in which it appears. Understanding the ambiguity of natural languages is considered an AI-hard problem. Computational problems like this are the central objectives of Artificial Intelligence (AI) and Natural Language Processing (NLP) because they aim to solve the epistemological question of how the mind works. It has been studied since the beginning of NLP, and today is a central topic of this discipline.

Semantics and Discourse Processing for Expressive TTS

In this paper we present ongoing work to produce an expressive TTS reader that can be used both in text and dialogue applications. The system has been previously used to read (English) poetry and it has now been extended to apply to short stories. The text is fully analyzed both at phonetic and phonological level, and at syntactic and semantic level. The core of the system is the Prosodic Manager which takes as input discourse structures and relations and uses this information to modify parameters for the TTS accordingly. The text is transformed into a poem-like structures, where each line corresponds to a Breath Group, semantically and syntactically consistent. Stanzas correspond to paragraph boundaries. Analogical parameters are related to ToBI theoretical indices but their number is doubled.

Linguistically Based QA by Dinamyc LOD Access from Logical Form

We present a system for Question Answering which computes a prospective answer from Logical Forms produced by a full-fledged NLP for text understanding, and then maps the result onto schemata in SPARQL to be used for accessing the Semantic Web. It is just by the internal structure of the Logical Form that we are able to produce a suitable and meaningful context for concept disambiguation. Logical Forms are the final output of a complex system for text understanding – VENSES which can deal with different levels of syntactic and semantic ambiguity in the generation of a final structure, by accessing computational lexical equipped with sub-categorization frames and appropriate selectional restrictions applied to the attachment of complements and adjuncts. The system also produces pronominal binding and instantiates the implicit arguments, if needed, in order to complete the required Predicate Argument structure which is licensed by the semantic component.

Linguistically-Based Reranking of Google’s Snippets with GreG

We present an experiment evaluating the contribution of a system called GReG for reranking the snippets returned by Google’s search engine in the 10 hits presented to the user and captured by the use of Google’s API. The evaluation aims at establishing whether or not the introduction of deep linguistic information may improve the accuracy of Google or rather it is the opposite case as maintained by the majority of people working in Information Retrieval and using a Bag Of Words approach. We used 900 questions and answers taken from TREC 8 and 9 competitions and execute three different types of evaluation: one without any linguistic aid; a second one with tagging and syntactic constituency contribution; another run with what we call Partial Logical Form. Even though GReG is still work in progress, it is possible to draw clear cut conclusions: adding linguistic information to the evaluation process of the best snippet that can answer a question improves enormously the performance. In another experiment we used the actual texts associated to the Q/A pairs distributed by one of TREC’s participant and got even higher accuracy.