Abstract
In recent years, TRIZ have been developed rapidly and known as a useful tool to solve the contradictions on engineering problems. The aim of this research is to map R.O.C patents into TRIZ applications by using text mining technique for further innovative product concepts and solutions of conflicts. First of all, using on-line auto-tag system provided by Academia Sinica to break every sentence in a document into several keywords and label these keywords manually. Calculating text frequency (TF) and inverse document frequency (IDF) in the corresponding document. Secondly, chi-square statistics and correlation coefficient approaches are used to select and sort word features which are highly correlated to 40 TRIZ innovative principles. Then, calculating the TFIDF and weight-TFIDF values for selected words, these values are inputs for further classifiers such as support vector machine (SVM) and BPN. Finally, SVM and BPN are implemented to evaluate the performances of 1000 R.O.C patents mapping into 40 TRIZ innovative principles. Experimental results show that, by appropriate parameters setting, SVM and BPN are accurate classifiers for 26 innovative principles (around 96%). However, SVM and BPN perform moderately for other 14 innovative principles (around 55%). Even though, this research still significantly help R&D personnel to solve trade-off contradictions by fast searching the correlation between TRIZ and existing R.O.C patents, which could reduce time-consuming for creative thinking and time-spent for new product introduction.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
