Anzeige
Erschienen in:
2015 | OriginalPaper | Buchkapitel
Knowledge Extraction from Web Reviews Using Feature Selection Based on Onomatopoeia
verfasst von : Fumiaki Saitoh, Hikaru Aoki, Shohei Ishizu
Erschienen in: HCI International 2015 - Posters’ Extended Abstracts
Verlag: Springer International Publishing
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
In the field of Buzz marketing, it is important to extract knowledge to improve products and services from the voice of the customer represented by customer reviews. In Japanese web review sentences, words that co-occur with onomatopoeia it has been confirmed that easy to combine with use sense of product. For sensory evaluation using a products can be easily associated with the satisfaction is obvious, onomatopoeia can be expected to contribute in knowledge extraction on customer satisfaction. A knowledge model for customer satisfaction is constructed by a regression tree that co-occurrence words with onomatopoeias are used as explanatory variables. Effectiveness of the proposed method I was confirmed through the analysis for the customer review data of ramen shop in Tokyo. The knowledge model acquired by our approach contained many words associated with noodles and food, on the other hand the normal regression tree model was included many meaningless words and unrelated words.