The purpose of this study is to clarify the mechanism of fea- ture emergence within the process o... more The purpose of this study is to clarify the mechanism of fea- ture emergence within the process of understanding metaphors of the form "A(target) like B(vehicle)" based on model sim- ulation and psychological experiments. Feature emergence refers to low-salient features of a target and a vehicle being em- phasized in the metaphor understanding process((Nueckles & Janetzko, 1997)(Becker, 1997)et al.). However,
International Conference on Natural Language Generation, 2010
In the field of referring expression generation, while in the static domain both intrinsic and ex... more In the field of referring expression generation, while in the static domain both intrinsic and extrinsic evaluations have been considered, extrinsic evaluation in the dynamic domain, such as in a situated collaborative dialog, has not been discussed in depth. In a dynamic domain, a crucial problem is that referring expressions do not make sense without an appropriate preceding dialog context. It is unrealistic for an evaluation to simply show a human evaluator the whole period from the beginning of a dialog up to the time point at which a referring expression is used. Hence, to make evaluation feasible it is indispensable to determine an appropriate shorter context. In order to investigate the context necessary to understand a referring expression in a situated collaborative dialog, we carried out an experiment with 33 evaluators and a Japanese referring expression corpus. The results contribute to finding the proper contexts for extrinsic evalution in dynamic domains.
The purpose of this research is to propose a development method of a metaphorical search engine t... more The purpose of this research is to propose a development method of a metaphorical search engine that reflects the metaphor understanding process of human-beings. Our goal is to realize a search engine which outputs the image of an object inputted metaphorically using the expression of "A like B". For example, when you input "A dog like clouds", the metaphorical search engine would output a visual image like the photograph of a white and puffy dog as the image. In this research, at first, the probabilistic relationship between concepts and their attributes is computed using a statistical analysis of language data to develop a computational model of metaphor understanding process. Secondly, psychological experiments are executed to examine the validity of the probabilistic relationship. Finally, using those results, a development method of a metaphorical search engine is proposed.
The purpose of the present study is to propose computational models of human inductive reasoning,... more The purpose of the present study is to propose computational models of human inductive reasoning, using a statistical analysis of Japanese linguistic data, and to develop a search- engine based on inductive reasoning. Osherson, et al. (1990) provided a psychological model of inductive reasoning based on the similarity between the premise and the conclusion and on knowledge of the category
The purpose of this study is to construct a computational model of metaphor understanding based o... more The purpose of this study is to construct a computational model of metaphor understanding based on statistical corpora analysis. The constructed model consists of two processes: a categorization process and a dynamic interaction process. The model expresses features based not only on adjectives but also on verbs using adjective-noun and three types of verb-noun modification data. The dynamic interaction is
... Page 2. 780 A. Terai and M. Nakagawa understand. ... Inputting a set of expressions (eg tran... more ... Page 2. 780 A. Terai and M. Nakagawa understand. ... Inputting a set of expressions (eg transienthope and hope disappear) to the model, it outputs an adequate noun for the Page 3. A Neural Network Model of Metaphor Generation with Dynamic Interaction 781 ...
The purpose of this study is to construct a computational model of the metaphor understanding pro... more The purpose of this study is to construct a computational model of the metaphor understanding process. This study assumes that metaphor understanding consists of two processes. The first is a categorization process; a target is assigned to an ad hoc category of which the vehicle is a prototypical member. The second is a dynamic interaction process; the target assigned to
... the system output the original metaphor hope like pin money as the most adequate metaphor f... more ... the system output the original metaphor hope like pin money as the most adequate metaphor for transient hope and hope disappears. That suggests that the system has the potential to generate more original metaphors than humans. References 1. Terai, A., Nakagawa, M ...
The purpose of this paper is to construct a model that represents the human process of understand... more The purpose of this paper is to construct a model that represents the human process of understanding metaphors, focusing specifically on similes of the form an "A like B". Generally speaking, human beings are able to generate and understand many sorts of metaphors. This study constructs the model based on a probabilistic knowledge structure for concepts which is computed from a statistical analysis of a large-scale corpus. Consequently, this model is able to cover the many kinds of metaphors that human beings can generate. Moreover, the model implements the dynamic process of metaphor understanding by using a neural network with dynamic interactions. Finally, the validity of the model is confirmed by comparing model simulations with the results from a psychological experiment.
The purpose of this study is to examine the understanding mechanism for metaphors represented in ... more The purpose of this study is to examine the understanding mechanism for metaphors represented in the form of “A is like B” (simile) using model simulation. In a previous experimental study[1] , the priming effects of visual images on metaphor processing were demonstrated in a psychological experiment. In that experiment, the presentation of a picture of a vehicle (”B”) interfered
The purpose of this study is to construct a probabilistic hierarchical structure of categories ba... more The purpose of this study is to construct a probabilistic hierarchical structure of categories based on a statistical analysis of Japanese corpus data and to verify the validity of the structure by conducting a psychological experiment. At first, the co-occurrence frequencies of adjectives and nouns within modification relations were extracted from a Japanese corpus. Secondly, a probabilistic hierarchical structure was
International Conference on Semantic Computing (ICSC 2007), 2007
The purpose of this study is to develop a method of constructing a probabilistic hierarchical str... more The purpose of this study is to develop a method of constructing a probabilistic hierarchical structure based on a statistical analysis of a Japanese corpus using a combination of Kameya and Sato's statistical language analysis and Rose's model. First, the co-occurrence frequencies of adjectives and nouns are calculated from a Japanese corpus based on modification relations. Second, latent classes are
Identifying objects in conversation is a fundamental human capability necessary to achieve effici... more Identifying objects in conversation is a fundamental human capability necessary to achieve efficient collaboration on any real world task. Hence the deepening of our understanding of human referential behaviour is indispensable for the creation of systems that collaborate with humans in a meaningful way. We present the construction of REX-J, a multi-modal Japanese corpus of referring expressions in situated dialogs, based on the collaborative task of solving the Tangram puzzle. This corpus contains 24 dialogs with over 4 h of recordings and over 1,400 referring expressions. We outline the characteristics of the collected data and point out the important differences from previous corpora. The corpus records extra-linguistic information during the interaction (e.g. the position of pieces, the actions on the pieces) in synchronization with the participants' utterances. This in turn allows us to discuss the importance of creating a unified model of linguistic and extra-linguistic information from a new perspective. Demonstrating the potential
Existing computational models of human inductive reasoning have been constructed based on psychol... more Existing computational models of human inductive reasoning have been constructed based on psychological evaluations concerning the similarities or relationships between entities. However, the costs involved in collecting psychological evaluations for the sheer number of entities that exist mean that they are prohibitively impractical. In order to avoid this problem, the present article examines three types of models: a category-based neural network model, a category-based Bayesian model, and a feature-based neural network model. These models utilize the results of a statistical analysis of a Japanese corpus computing co-occurrence probabilities for word pairs, rather than using psychological evaluations. Argument strength ratings collected by a psychological experiment were found to correlate well with simulations for the category-based neural network model.
Emergence of creative ideas often involves sudden linking of unrelated ideas. Here we demonstrate... more Emergence of creative ideas often involves sudden linking of unrelated ideas. Here we demonstrate that the production of insights about novel metaphor was preceded by prolonged visual inspection of metaphorical sentences. The participants made free spoken responses in a task requiring the generation of semantic interpretations about visually presented sentences, where gaze locations were continuously monitored. We found that creative interpretations were primarily generated from novel metaphorical expressions, rather than from conventional expressions. Moreover, presented words within novel metaphors were visually inspected for longer periods specifically before creative interpretations were produced. Interestingly, prolonged gazes occurred several seconds (∼8 s) prior to the generation of creative interpretations, particularly, in the case of the topic word within the metaphor. These results demonstrate latent cognitive process meditating the emergence of insights, and suggest that the prior visual inspection prompted the semantic representation of the metaphorical sentences, which eventually facilitates creative ideas.
The purpose of this study is to clarify the mechanism of fea- ture emergence within the process o... more The purpose of this study is to clarify the mechanism of fea- ture emergence within the process of understanding metaphors of the form "A(target) like B(vehicle)" based on model sim- ulation and psychological experiments. Feature emergence refers to low-salient features of a target and a vehicle being em- phasized in the metaphor understanding process((Nueckles & Janetzko, 1997)(Becker, 1997)et al.). However,
International Conference on Natural Language Generation, 2010
In the field of referring expression generation, while in the static domain both intrinsic and ex... more In the field of referring expression generation, while in the static domain both intrinsic and extrinsic evaluations have been considered, extrinsic evaluation in the dynamic domain, such as in a situated collaborative dialog, has not been discussed in depth. In a dynamic domain, a crucial problem is that referring expressions do not make sense without an appropriate preceding dialog context. It is unrealistic for an evaluation to simply show a human evaluator the whole period from the beginning of a dialog up to the time point at which a referring expression is used. Hence, to make evaluation feasible it is indispensable to determine an appropriate shorter context. In order to investigate the context necessary to understand a referring expression in a situated collaborative dialog, we carried out an experiment with 33 evaluators and a Japanese referring expression corpus. The results contribute to finding the proper contexts for extrinsic evalution in dynamic domains.
The purpose of this research is to propose a development method of a metaphorical search engine t... more The purpose of this research is to propose a development method of a metaphorical search engine that reflects the metaphor understanding process of human-beings. Our goal is to realize a search engine which outputs the image of an object inputted metaphorically using the expression of "A like B". For example, when you input "A dog like clouds", the metaphorical search engine would output a visual image like the photograph of a white and puffy dog as the image. In this research, at first, the probabilistic relationship between concepts and their attributes is computed using a statistical analysis of language data to develop a computational model of metaphor understanding process. Secondly, psychological experiments are executed to examine the validity of the probabilistic relationship. Finally, using those results, a development method of a metaphorical search engine is proposed.
The purpose of the present study is to propose computational models of human inductive reasoning,... more The purpose of the present study is to propose computational models of human inductive reasoning, using a statistical analysis of Japanese linguistic data, and to develop a search- engine based on inductive reasoning. Osherson, et al. (1990) provided a psychological model of inductive reasoning based on the similarity between the premise and the conclusion and on knowledge of the category
The purpose of this study is to construct a computational model of metaphor understanding based o... more The purpose of this study is to construct a computational model of metaphor understanding based on statistical corpora analysis. The constructed model consists of two processes: a categorization process and a dynamic interaction process. The model expresses features based not only on adjectives but also on verbs using adjective-noun and three types of verb-noun modification data. The dynamic interaction is
... Page 2. 780 A. Terai and M. Nakagawa understand. ... Inputting a set of expressions (eg tran... more ... Page 2. 780 A. Terai and M. Nakagawa understand. ... Inputting a set of expressions (eg transienthope and hope disappear) to the model, it outputs an adequate noun for the Page 3. A Neural Network Model of Metaphor Generation with Dynamic Interaction 781 ...
The purpose of this study is to construct a computational model of the metaphor understanding pro... more The purpose of this study is to construct a computational model of the metaphor understanding process. This study assumes that metaphor understanding consists of two processes. The first is a categorization process; a target is assigned to an ad hoc category of which the vehicle is a prototypical member. The second is a dynamic interaction process; the target assigned to
... the system output the original metaphor hope like pin money as the most adequate metaphor f... more ... the system output the original metaphor hope like pin money as the most adequate metaphor for transient hope and hope disappears. That suggests that the system has the potential to generate more original metaphors than humans. References 1. Terai, A., Nakagawa, M ...
The purpose of this paper is to construct a model that represents the human process of understand... more The purpose of this paper is to construct a model that represents the human process of understanding metaphors, focusing specifically on similes of the form an "A like B". Generally speaking, human beings are able to generate and understand many sorts of metaphors. This study constructs the model based on a probabilistic knowledge structure for concepts which is computed from a statistical analysis of a large-scale corpus. Consequently, this model is able to cover the many kinds of metaphors that human beings can generate. Moreover, the model implements the dynamic process of metaphor understanding by using a neural network with dynamic interactions. Finally, the validity of the model is confirmed by comparing model simulations with the results from a psychological experiment.
The purpose of this study is to examine the understanding mechanism for metaphors represented in ... more The purpose of this study is to examine the understanding mechanism for metaphors represented in the form of “A is like B” (simile) using model simulation. In a previous experimental study[1] , the priming effects of visual images on metaphor processing were demonstrated in a psychological experiment. In that experiment, the presentation of a picture of a vehicle (”B”) interfered
The purpose of this study is to construct a probabilistic hierarchical structure of categories ba... more The purpose of this study is to construct a probabilistic hierarchical structure of categories based on a statistical analysis of Japanese corpus data and to verify the validity of the structure by conducting a psychological experiment. At first, the co-occurrence frequencies of adjectives and nouns within modification relations were extracted from a Japanese corpus. Secondly, a probabilistic hierarchical structure was
International Conference on Semantic Computing (ICSC 2007), 2007
The purpose of this study is to develop a method of constructing a probabilistic hierarchical str... more The purpose of this study is to develop a method of constructing a probabilistic hierarchical structure based on a statistical analysis of a Japanese corpus using a combination of Kameya and Sato's statistical language analysis and Rose's model. First, the co-occurrence frequencies of adjectives and nouns are calculated from a Japanese corpus based on modification relations. Second, latent classes are
Identifying objects in conversation is a fundamental human capability necessary to achieve effici... more Identifying objects in conversation is a fundamental human capability necessary to achieve efficient collaboration on any real world task. Hence the deepening of our understanding of human referential behaviour is indispensable for the creation of systems that collaborate with humans in a meaningful way. We present the construction of REX-J, a multi-modal Japanese corpus of referring expressions in situated dialogs, based on the collaborative task of solving the Tangram puzzle. This corpus contains 24 dialogs with over 4 h of recordings and over 1,400 referring expressions. We outline the characteristics of the collected data and point out the important differences from previous corpora. The corpus records extra-linguistic information during the interaction (e.g. the position of pieces, the actions on the pieces) in synchronization with the participants' utterances. This in turn allows us to discuss the importance of creating a unified model of linguistic and extra-linguistic information from a new perspective. Demonstrating the potential
Existing computational models of human inductive reasoning have been constructed based on psychol... more Existing computational models of human inductive reasoning have been constructed based on psychological evaluations concerning the similarities or relationships between entities. However, the costs involved in collecting psychological evaluations for the sheer number of entities that exist mean that they are prohibitively impractical. In order to avoid this problem, the present article examines three types of models: a category-based neural network model, a category-based Bayesian model, and a feature-based neural network model. These models utilize the results of a statistical analysis of a Japanese corpus computing co-occurrence probabilities for word pairs, rather than using psychological evaluations. Argument strength ratings collected by a psychological experiment were found to correlate well with simulations for the category-based neural network model.
Emergence of creative ideas often involves sudden linking of unrelated ideas. Here we demonstrate... more Emergence of creative ideas often involves sudden linking of unrelated ideas. Here we demonstrate that the production of insights about novel metaphor was preceded by prolonged visual inspection of metaphorical sentences. The participants made free spoken responses in a task requiring the generation of semantic interpretations about visually presented sentences, where gaze locations were continuously monitored. We found that creative interpretations were primarily generated from novel metaphorical expressions, rather than from conventional expressions. Moreover, presented words within novel metaphors were visually inspected for longer periods specifically before creative interpretations were produced. Interestingly, prolonged gazes occurred several seconds (∼8 s) prior to the generation of creative interpretations, particularly, in the case of the topic word within the metaphor. These results demonstrate latent cognitive process meditating the emergence of insights, and suggest that the prior visual inspection prompted the semantic representation of the metaphorical sentences, which eventually facilitates creative ideas.
Uploads
Papers by Asuka Terai