Fig 1.
Concept of identifying needs in vector space.
Underlined word, blue color word and red colored word expressing sentiment, object and function, respectively.
Fig 2.
Module construction (outer flow chart), detailed processes for each module (inner flow chart), and methodologies (tagged by right side boxes).
Fig 3.
Identifying needs using combination of trigger (example).
Underlined word, bold word and italics word expressing sentiment, object and function, respectively.
Table 1.
Identifying needs using needs formula.
Fig 4.
Process to calculate sentimental value of needs.
A conceptual example of needs with SK and OT (left box) and flow chart of calculating s-value based on SK and OT (right flow chart).
Fig 5.
Process of identify needs-related technology.
A conceptual example of needs with OT and related words (left rounded box), needs with trigger and related word (middle top table), patent list with related OT word lists (middle bottom table) and need list with related patents (right table).
Table 2.
Meaning of patent index and formula.
Table 3.
Four types of needs.
Table 4.
Keyword frequency in review data.
Table 5.
Needs ratio based on keywords and documents.
Table 6.
Extracted sentimental keywords from opinions.
Table 7.
Identified opinion triggers (part).
Table 8.
Discovered needs of users (part).
Table 9.
Sentimental value of needs for candidate technology opportunity.
Table 10.
Needs-technology relationship and technology ability.
Table 11.
Classified needs and technology opportunity.
Table 12.
Property of technology opportunity.
Table 13.
Cluster type based on DTM & K-means clustering.
Table 14.
ODI complaints data related to the needs.
Table 15.
Machine learning (Word2vec + PAM clustering) accuracy.
Table 16.
Annual growth rate of needs.
Table 17.
Annual patent growth rate of needs.