HanLP-Based Technology Function Matrix Construction on Chinese Process Patents

HanLP-Based Technology Function Matrix Construction on Chinese Process Patents

Yuexin Yang (Shaanxi University of Science and Technology, China) and Gongchang Ren (Shaanxi University of Science and Technology, China)
DOI: 10.4018/IJMCMC.2020070104
OnDemand PDF Download:
No Current Special Offers


The technology/function matrix is an essential method of patent analysis. This paper presents a semi-automated methodology to create technology/function matrix for Chinese process patents in any given domain. The approach adopts computer-aided algorithms in the process of extracting technology words, extracting function words, and generating technology/function matrix. The purpose of this paper is to increase the efficiency of building technology/function matrix and reduce the cost of manpower and time. The Chinese process patents in the domain of robot path planning art taken as an example to verify the methodology.
Article Preview

1. Introduction

The purpose of patent analysis is to provide reference to the decision-making in the development of technology, products and services (Kim G, 2017; Kim D et al., 2018). The process of patent analysis is to analyze, process and combine a large amount of unstructured information on patent documents, and use statistical methods to get valuable technical information (e.g identification of technological opportunities, technology trends, etc.) (Yoon J, 2017; Sampaio, P. G. V. et al., 2018; Moehrle, M. G. et al., 2019). The technologies of data mining (discriminant analysis, rule induction, visualization technology, etc.) are introduced into patent analysis to liberate people from the monotonous and boring work of reading patent documents, enhance the efficiency and accuracy of analysis (Lee, J. H. et al., 2018; Kim, J. M. et al., 2019; Son, C, 2018). Patent analysis also belongs to Knowledge Discovery in Databases. In the process of patent analysis, a variety of types of knowledge and expertise are needed (Zemmouri, E. M. et al., 2012). The results of patent analysis can show the hidden relations and laws in patent documents by means of graphical means in multimedia technology, which makes it easier for researchers to understand and make decisions based on them.

The technology/function matrix is a commonly eminent method of patent analysis. The technology/function matrix is usually posted in the form of bubble graph. In the bubble graph, the x axis represents a function, the y axis represents technology, the bubble size represents the quantity of patents . The cross point having no bubble may be a technology gap in a given domain. The technical/function matrix can help managers, researchers and engineers take a decision on the technological development in a given domain.

The construction process flow of technology/function matrix is depicted in Figure 1.

Figure 1.

Construction process of technology/function Matrix


The general process of creating a technical/function matrix consist of three steps:

  • The first step consists of three processes: searching for invention patents in a given domain by keyword, screening invention patents, storing information about invention patents;

  • The second step include two processes: extracting technology words of the process patents and literature review in the given domain, extracting function words of the process patents;

  • The third step is composed of three processes: tagging technology words of invention patents, tagging function words of invention patents, generating the technology/function matrix.

In the general process of creating a technical/function matrix, it goes to get a lot of time to extract technology words and function words. The abstract of invention patents needs to be analyzed manually in order to tag the technology words and function words used in the process patent.

Complete Article List

Search this Journal:
Volume 14: 1 Issue (2023): Forthcoming, Available for Pre-Order
Volume 13: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 12: 4 Issues (2021)
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing