This paper presents a new approach for measuring semantic similarity/distance between words and concepts. It combines a lexical taxonomy structure with corpus statistical information so that the semantic distance between nodes in the semantic space constructed by the taxonomy can be better quantified with the computational evidence derived from a distributional analysis of corpus data. Specifically, the proposed measure is a combined approach that inherits the edge-based approach of the edge counting scheme, which is then enhanced by the node-based approach of the information content calculation. When tested on a common data set of word pair similarity ratings, the proposed approach outperforms other computational models. It gives the highest correlation value (r = 0.828) with a benchmark based on human similarity judgements, whereas an upper bound (r = 0.885) is observed when human subjects replicate the same task.

/pdf/semantic-similarity-based-on-corpus-statistics-and-lexical-58z85g615l.pdf

Semantic Similarity Based on Corpus Statistics and Lexical Taxonomy

https://www.cs.cmu.edu/~illah/PAPERS/socialroboticssurvey.pdf

A survey of socially interactive robots

The cultural origins of human cognition

The quantification of lexical semantic relatedness has many applications in NLP, and many different measures have been proposed. We evaluate five of these measures, all of which use WordNet as their central resource, by comparing their performance in detecting and correcting real-word spelling errors. An information-content-based measure proposed by Jiang and Conrath is found superior to those proposed by Hirst and St-Onge, Leacock and Chodorow, Lin, and Resnik. In addition, we explain why distributional similarity is not an adequate proxy for lexical semantic relatedness.

/pdf/evaluating-wordnet-based-measures-of-lexical-semantic-3o17unbmj1.pdf

Evaluating WordNet-based Measures of Lexical Semantic Relatedness

TextTiling is a technique for subdividing texts into multi-paragraph units that represent passages, or subtopics. The discourse cues for identifying major subtopic shifts are patterns of lexical co-occurrence and distribution. The algorithm is fully implemented and is shown to produce segmentation that corresponds well to human judgments of the subtopic boundaries of 12 texts. Multi-paragraph subtopic segmentation should be useful for many text analysis tasks, including information retrieval and summarization.

/pdf/texttiling-segmenting-text-into-multi-paragraph-subtopic-63iqujn0hq.pdf

TextTiling: segmenting text into multi-paragraph subtopic passages

Keepon is a small creature-like robot designed for simple, natural, nonverbal interaction with children. The minimal design of Keepon’s appearance and behavior is meant to intuitively and comfortably convey the robot’s expressions of attention and emotion. For the past few years, we have been observing interactions between Keepon and children at various levels of physical, mental, and social development. With typically developing children, we have observed varying styles of play that suggest a progression in ontological understanding of the robot. With children suffering from developmental disorders such as autism, we have observed interactive behaviors that suggest Keepon’s design is effective in eliciting a motivation to share mental states. Finally, in developing technology for interpersonal coordination and interactional synchrony, we have observed an important role of rhythm in establishing engagement between people and robots. This paper presents a comprehensive survey of work done with Keepon to date.

Keepon: A playful robot for research, therapy, and entertainment

The paper proposes possible use of interactive robots in the remedial practice for children with autism, who have difficulties mainly in interpersonal communication. For this purpose, we built a small creature-like robot, Keepon, which was carefully designed to get autistic and non-autistic children involved in playful interaction. We observed how autistic children (2-4 years old) interacted with Keepon without any experimental setting or instructions in a playroom at a day-care center for children with special needs. From the longitudinal observation for a year and a half (totally, over 500 child-sessions), we found that Keepon's simple appearance and predictable responses gave the autistic children a playful and relaxed mood, in which they spontaneously engaged in dyadic play with Keepon, which would then expand into interpersonal communication where Keepon worked as the pivot of triadic play with adults or other children. Each child showed a different style and a different unfolding of interaction over time, which tell us a "story" of his or her personality and developmental profile, which would not be explained thoroughly by a diagnostic label like "autism".

Interactive robots for communication-care: a case-study in autism therapy

We present here a pilot study of child-robot interactions, in which we discuss developmental origins of human interpersonal communication. For the past few years, we have been observing 2- to 4-year-old children with autism interacting with Keepon, a creature-like robot that is only capable of expressing its attention (directing its gaze) and emotions (pleasure and excitement). While controlled by a remote experimenter, Keepon interacted with the children with its simple appearance and actions. With a sense of curiosity and security, the children spontaneously approached Keepon and engaged in dyadic interaction with it, which then extended to triadic interactions where they exchanged with adult caregivers pleasure and surprise they found in Keepon. Qualitative and quantitative analysis of these unfolding interactions suggests that autistic children possess the motivation to share mental states with others, which is contrary to the commonly held position that this motivation is impaired in autism. We assume Keepon's minimal expressiveness helped the children understand socially meaningful information, which then activated their intact motivation to share interests and feelings with others. We conclude that simple robots like Keepon would facilitate social interaction and its development in autistic children.

Children-robot interaction: a pilot study in autism therapy.

This paper proposes a new indicator of text structure, called the lexical cohesion profile (LCP), which locates segment boundaries in a text. A text segment is a coherent scene; the words in a segment are linked together via lexical cohesion relations. LCP records mutual similarity of words in a sequence of text. The similarity of words, which represents their cohesiveness, is computed using a semantic network. Comparison with the text segments marked by a number of subjects shows that LCP closely correlates with the human judgments. LCP may provide valuable information for resolving anaphora and ellipsis.

Hideki Kozima

Papers

Keepon: A playful robot for research, therapy, and entertainment

Interactive robots for communication-care: a case-study in autism therapy

Children-robot interaction: a pilot study in autism therapy.

Text segmentation based on similarity between words

Text Segmentation Based on Similarity between Words