Search

Increase in computer performance in recent years enables us to obtain a variety of valuable knowledge from original data without structurizing text data. Text-mining is expected to play an important role in knowledge acquisition more and more in future. For instance, it will be useful for the management of technology (MOT) including decision making for "selection and concentration" strategy. However, the technique of text-mining for discovering knowledge, with the reliability which is necessary for decision making, has not yet fully been established. Statistical analysis is insufficient for acquiring knowledge trusted by managers. Therefore, in this article, authors verify whether human-computer collaboration, in which human-embodied knowledge is utilized for text mining, can be useful for MOT. To do so, taking an example of quantum dot solar cell, authors argue that "word sets" , in which words are selected carefully and are put together through human-computer collaboration, are useful for discovering knowledge meaningful for decision-making by companies who deal with huge amounts of text data. [ABSTRACT FROM AUTHOR]