Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/5672B4FE-6BAF-436C-A990-B7383F90556B | ||
| Year | 2019 | ||
| Type of Item | Master Thesis | ||
| License |
|
||
| Bibliographic Citation | Aikaterini Nerantzaki, "Use of neuro - fuzzy techniques for patent prediction", Master Thesis, School of Production Engineering and Management, Technical University of Crete, Chania, Greece, 2019 https://doi.org/10.26233/heallink.tuc.81276 | ||
| Appears in Collections |
Recent innovations in processing power and digitisation of patent data has allowed a massive access to databases, which was used to conduct the current research. The approach used in the research, belongs to the field of predictive analysis; an area of data analysis, which deals with the anticipation of future trends. The main element of predictive analysis is its factor; a mathematical object which can be defined for an individual, organisation, or other entity, that can be used to predict its future behaviour. The current thesis focuses on the attempt to predict patents submitted in China daily; taking into consideration the number of patents submitted between 30/10/2014 and 15/01/2019. The method used to extract information regarding the number of patents in specific time and place is known as Anfis (Adaptive Neuro Fuzzy Inference System). The Anfis receives as input the daily values and as the output the prediction for the patent values. The aim of the thesis is to provide patents for the following few days with the highest possible precision and the minimum possible error.
Copyright © DIAS 2013
