Non-Hierarchichal Network Evolutionary System

July 28, 2010

CEC 2010: Evolving Natural Language Grammars without Supervision

Filed under: Uncategorized — lurdesaraujo @ 8:57 am

Another work presented world congress on computational intelligence has been: Evolving Natural Language Grammars without Supervision. paper: Training a Classifier for Good Query Expansion Terms with a Genetic Algorithm.

This work is devoted to unsupervised grammar induction, whose goal is to extract a grammar representing the language structure using texts without annotations of this structure. We have devised an evolutionary algorithm which for each sentence evolves a population of trees that represent different parse trees of that sentence. Each of these trees represent a part of a grammar.
The evaluation function takes into account the contexts in which each sequence of Part-Of-Speech tags (POSseq) appears in the training corpus, as well as the frequencies of those POSseqs and contexts.
The algorithm has been evaluated using a well known Annotated English corpus, though the annotation have only been used for evaluation purposes. Results indicate that the proposed algorithm is able to improve
the results of a classical optimization algorithm, such as EM (Expectation Maximization), for short
grammar constituents (right side of the grammar rules), and its precision is better in general.

The presentation can be found in:.
The paper is not available yet at IEEEExplore, but we can email to whoever is interested.

Advertisements

July 26, 2010

cec 2010: good query expansion terms classifier

Filed under: Applications, Congreso, Genetic Algorithms — Tags: — lurdesaraujo @ 5:54 pm

This is the presentation of another world congress on computational intelligence paper: Training a Classifier for Good Query Expansion Terms with a Genetic Algorithm. This is about automatically finding good terms to improve the queries the users submit to the search systems, and this is the presentation: .
We have developed a classifier which has been trained for distinguishing good expansion terms. The identification of good terms to train the classifier has been achieved with a genetic algorithm whose fitness
function is based on users’ relevance judgements on a set of documents. Results show that the training performed by the genetic algorithm is able to improve the quality of the query expansion results.
The paper is not available yet at IEEEExplore, but we can email to whoever is interested.

July 25, 2010

Fluid Evolutionary Algorithms at WCCI’2010

Filed under: Uncategorized — Tags: , , — jjmerelo @ 5:04 pm

We are freshly back from the world congress on computational intelligence, and papers do not seem to be available yet. However, our presentations are. This is a paper on Fluid Evolutionary Algorithms, basically figuring out how a more or less traditional EA could be implemented over the new FluidDB, a database with the spirit of a wiki.

The paper is not available yet at IEEEExplore, but we can email to whoever is interested.

Blog at WordPress.com.