Analysis of the effect of data properties in automated patent classification

Abstract

Patent classification is a task performed in patent offices around the world by experts, where they assign category codes to a patent application based on its technical content. Nowadays, the number of applications is constantly growing and there is an economical interest on developing accurate and fast models to automate the classification task. In this paper, we present a methodology to systematically analyze the effect of three patent data properties and two classification details on the patent classification task: patent section to use for training/testing, document representation, patent codes to use for training, use of the hierarchy of categories, and the base classifier. For the analysis we create a diversity of models by combining different options for the properties. We evaluate the models in detail using standard patent datasets in two languages, English and German, considering three performance metrics, using statistical tests to validate the results and comparing them with other models in the literature. Our research findings indicate that it is important to follow a methodology to properly choose the options for the data properties to build a model according to our goal, considering classification accuracy and computational efficiency. Some combinations of options build models with good results but with high computational cost, whilst other build model that produce slightly worst results but at a fraction of the training time.

Semantic similarity measure of natural language text through machine learning and a keyword‐aware cross‐encoder‐ranking summarizer—A case study using UCGIS GIS&T body of knowledge

“…Semantic similarity is useful in various NLP tasks such as information retrieval, machine translation, question answering, and entity resolution (Ebraheem et al, 2018; Li et al, 2019; Varelas et al, 2005; Yih et al, 2014; Zou et al, 2013). In practice, it has a wide range of real‐life applications such as patent class prediction, scientific text comparison, newspaper article similarity analysis, topic discovery for United Nations speeches, and concept derivation for encyclopedias (Geum & Kim, 2020; Gomez, 2019; Gong et al, 2019; Shalaby & Zadrozny, 2017; Watanabe & Zhou, 2020). Going beyond determining the semantic similarity as a binary decision (1 or 0, as similar or not), generating non‐binary scores of text pairs followed by ranking has become a common goal of semantic similarity analysis.…”

Data-Driven Technology Roadmaps to Identify Potential Technology Opportunities for Hyperuricemia Drugs

“…Data analysis tools, such as text classification models, can be used to put the data source selected into proper layers of data-driven TRM. Classification models such as support vector machine (SVM) [86], k-nearest neighbor (KNN) [87,88], Hidden Markov [89], and Bayesian [44,90] can be employed.…”

PatentNet: multi-label classification of patent documents using deep learning based language understanding

“…Some focused on the best way to represent the patent text and how to extract semantic features from it (D'hondt et al, 2013;Shalaby et al, 2018;Hu et al, 2018a;Hu et al, 2018b;Li et al, 2018) while others focused on designing more effective classification algorithms (Fall et al, 2003;Al Shamsi & Aung, 2016;D'hondt et al, 2017;Wu et al, 2010Wu et al, , 2016Song et al, 2019). Furthermore, some attempts have been made to find which part of the patent text can be more representative and provide better classification results (Gomez, 2019;Hu et al, 2018a;Wu et al, 2010;D'hondt & Verberne, 2010). Gomez & Moens (2014) did a comprehensive survey of several previous works that tackled the automated patent classification problem in the IPC hierarchy.…”