Title page for etd-0627118-101540


[Back to Results | New Search]

URN etd-0627118-101540
Author Wei-En Huang
Author's Email Address cksh11233@gmail.com
Statistics This thesis had been viewed 5342 times. Download 0 times.
Department Information Management
Year 2017
Semester 2
Degree Master
Type of Document
Language English
Title Using web search data as market demand to assess patent value
Date of Defense 2018-07-20
Page Count 46
Keyword
  • patent information
  • word2vector
  • patent analysis
  • patent value
  • search data analysis
  • Abstract Assessing the value of patents is an important and difficult task in patent analysis. Some prior studies relied on experts to assess the value of patents, but it is time and labor consuming. As a result, recent studies tried to use some proxy variable to represent the patent value such as the binary occurrence of an opposition for the patent, the patent auction price, and the patent lifetime. In this study, we want to stand on the demand side to estimate the patent value. We believe that the attention to the knowledge and techniques of people can reflect the importance and the value of these knowledge and techniques and we call the degree of the attention as social voice. We apply search data analysis to calculate the social voice. The search data is from Google Trend. We extract the noun phrases of patents as the key knowledge and techniques and train a word2vector model to aggregate the similar semantic words. Then, we used these noun phrases as keywords to fetch the search volume and do further calculation and finally we get the two types of social voice, the total type and the trend type. We discuss two types of datasets, the technology-based dataset and the company-based dataset. In these datasets, we observe the relationship between our social voice and the patent information and compare our results with the prior studies. The results show that the relationship between total type social voice and patent information is similar to the prior studies and the relationship between trend type social voice and patent information is opposite to the prior studies. In conclusion, it is possible to use the total type social voice to assess the value of patent and the trend type social voice presents a whole new perspective to discuss the value of patent.
    Advisory Committee
  • Keng-Pei Lin - chair
  • Huang-Chih Sung - co-chair
  • Yi-Ling Lin - advisor
  • Files
  • etd-0627118-101540.pdf
  • Indicate in-campus at 5 year and off-campus access at 5 year.
    Date of Submission 2018-07-27

    [Back to Results | New Search]


    Browse | Search All Available ETDs

    If you have more questions or technical problems, please contact eThesys