Text categorization methods for automatic estimation of verbal intelligence

Abstract

In this paper we investigate whether conventional text categorization methods may suffice to infer different verbal intelligence levels. This research goal relies on the hypothesis that the vocabulary that speakers make use of reflects their verbal intelligence levels. Automatic verbal intelligence estimation of users in a spoken language dialog system may be useful when defining an optimal dialog strategy by improving its adaptation capabilities. The work is based on a corpus containing descriptions (i.e. monologs) of a short film by test persons yielding different educational backgrounds and the verbal intelligence scores of the speakers. First, a one-way analysis of variance was performed to compare the monologs with the film transcription and to demonstrate that there are differences in the vocabulary used by the test persons yielding different verbal intelligence levels. Then, for the classification task, the monologs were represented as feature vectors using the classical TF-IDF weighting scheme. The Naive Bayes, /c-nearest neighbors and Rocchio classifiers were tested. In this paper we describe and compare these classification approaches, define the optimal classification parameters and discuss the classification results obtained.

e-Dictionaries-based Semantic Gradient: Assisting Preschool Children Connect between Known and New Vocabulary

“…They were gathered to learn English. The use of this kind of sampling, which is mostly branded as convenience sampling, haphazard sampling, grab sampling, accidental sampling, incidental sampling, or opportunity sampling, was due to the relative ease of access to the research participants (subjects) because the subjects were readily available, expediting the data collection, saving times, and research expenses or cost-effectiveness (Christensen et al, 2012;Fernández-Martínez et al, 2012;Given, 2008;Henry, 1990;Teddlie & Yu, 2007;Wright, Julius Sim, 2002). This research lasted for 6 months while the learning process took place three times a week where each meeting lasted for 120 minutes.…”

I Feel You: The Design and Evaluation of a Domotic Affect-Sensitive Spoken Conversational Agent

“…Within the field of ambient intelligence (AmI), researchers are looking for ways to make agents adapt to users on the “fly”, either to the latter's emotions (as is the concern of this paper), preferences [2,3], age [4], environmental health [5] or, even, IQ [6]. The motivation is that an agent that adapts better to users would hopefully lead a more effective, empathetic and naturalistic interaction.…”

Research on Automatic Text Classification Algorithm Based on ITF-IDF and KNN

“…Automatic text classification is an important data mining methods that can bring these complex data structure into standard data structure, provide convenience to the user in retrieve useful knowledge or text efficiently and fastly. Therefore, it can be widely applied to information retrieval, digital library, information filtering, and search engine, among others [2].…”