A Computational Model of the Metaphor Generation Process

A Computational Model of the Metaphor Generation Process Keiga Abe (abe@nm.hum.titech.ac.jp) Kayo Sakamoto (sakamoto@nm.hum.titech.ac.jp) Masanori Nakagawa (nakagawa@nm.hum.titech.ac.jp) Graduate School of Decision Science & Technology, Tokyo Institute of Technology 2-12-1, Ohkayama, Meguro-Ku, Tokyo, 152-8552, Japan metaphor generation process was constructed based on the results of the statistical analysis. After that, a psy- chological experiment was conducted to examine the va- lidity of the model. Abstract The purpose of this research was to construct a com- putational model of the metaphor generation process. In order to construct the model, first, the probabilistic relationship between concepts and words was computed with a statistical analysis of language data. Secondly, a computational model of the metaphor generation pro- cess was constructed with results of the statistical anal- ysis of language data. The results of the simulation were examined from a comparison with metaphors that par- ticipants had generated. Finally, a third-party rating of the metaphors the model generated was conducted. Probabilistic representation of meaning In previous studies, practical methods to compute the probabilistic relationship between concepts and their words, between words and words have been developed. For example, LSA(Landauer & Dumais, 1997) assumes semantically similar words occur in common contexts. In LSA, text data are represented as a matrix in which each row stands for a unique word and each column for a text passage or other context. Each cell stands for the frequency with which the word of its row appears in the passage denoted by its column. After that, LSA applies singular value decomposition (SVD) to the matrix, as follows: S = U k Σ k U k . Introduction Metaphor understanding and generation processes are very important aspects of language study. How- ever, most cognitive studies of metaphor focus on the metaphor understanding process(Lakoff & Johnson, 1986; Glucksberg & Keysar, 1990; Kusumi, 1995), while studies of the metaphor generation processes are rela- tively few. The purpose of this study is to construct a computational model which generates a “A like B” style metaphor process. In the case of “A like B” sytle metaphors, A is called the “vehicle”, and B is called the “topic”. In a previous study, Kusumi(2003) showed that be- lief or experience affects the metaphor generating pro- cess, using a metaphor generation task dealing with the concept of love. Hisano(1996) studied the relationship between the impression of the topic and that of gener- ated metaphors, using a metaphor generation task where the categories of topic and vehicle were limited. How- ever, these studies were limited to a few concepts or cat- egories. It is not clear whether the results are applicable in the case of other concepts. In order to examine the applicability of the studies, the experimenter must con- duct a metaphor generation task with a huge number of concepts. It is impossible to cover large scale language knowledge using only a psychological experiment, be- cause psychological experiments require expensive time and labor. In order to solve this problem, a statistical analysis of language data was used to represent large scale human language knowledge stochastically. Applying statistical analysis, a stochastic language knowledge structure can be automatically constructed without subjective judge- ment. In this study, a statistical analysis of language data was conducted and a computational model of the Using this method, the meaning of words can be repre- sented in the coordinate of a vector space. Furthermore, semantic similarities between words and words are rep- resented by the cosine distance of vectors. However, LSA can not treat functional words(for ex- ample, “the”, “a”, “is”). Generally, functional words oc- cur in various contexts with high occurrence frequency. Such cooccurrence between content words and functional words do not necessarily reflect semantic relation. In order to avoid this problem, LSA has to set a strong weight to high occurrence frequency words. or omit low occurrence frequency words, However such a weighting method is likely to be subjective and ad-hoc. PLSI(Hofmann, 1999) is a probabilistic model for the relationship between concepts and words based on the idea of LSA. PLSI assumes that latent semantic classes c’s mediate the probability of cooccurrence between doc- uments d’s and words w’s. In PLSI, the probability of cooccurrence between a document d and a word w, P (d, w) is represented by the following equation: P (d|c)P (w|c)P (c), P (d, w) = c∈C where P (d|c) stands for the conditional probability of a document d, given a latent semantic class c, P (w|c) stands for the conditional probability of w, given c, and P (c) stands for the probability of c. Applying this