Classifying Patent Applications with Ensemble Methods

Classifying Patent Applications with Ensemble Methods Fernando Benites1, Shervin Malmasi2,3, Marcos Zampieri4 1Zurich University of Applied Sciences, Switzerland 2Harvard Medical School, United States 3Macquarie University, Australia 4University of Wolverhampton, United Kingdom benf@zhaw.ch, shervin.malmasi@mq.edu.au, m.zampieri@wlv.ac.uk Abstract We present methods for the automatic classification of patent applications using an annotated dataset provided by the or- ganizers of the ALTA 2018 shared task - Classifying Patent Applications. The goal of the task is to use computational methods to categorize patent applications accord- ing to a coarse-grained taxonomy of eight classes based on the International Patent Classification (IPC). We tested a variety of approaches for this task and the best results, 0.778 micro-averaged F1-Score, were achieved by SVM ensembles using a combination of words and characters as features. Our team, BMZ, was ranked first among 14 teams in the competition. 1 Introduction According to statistics of the World Intellectual Property Organization (WIPO),1 the number of patent applications filled across the world keeps growing every year. To cope with the large volume of applications, companies and organizations have been investing in the development of software to process, store, and categorize patent applications with minimum human intervention. An important part of patent application forms is, of course, composed of text. This has led to the widespread use of NLP methods in patent applica- tion processing systems as evidenced in Section 2. One such example is the use of text classification methods to categorize patent applications accord- ing to standardized taxonomies such as the Inter- national Patent Classification (IPC)2 as discussed in the studies by Benzineb and Guyot (2011); Fall et al. (2003). 1http://www.wipo.int/ipstats/en/ 2http://www.wipo.int/classifications/ ipc/en/ In this paper, we present a system to automati- cally categorize patent applications from Australia according to the top sections of the IPC taxonomy using a dataset provided by the organizers of the ALTA 2018 shared task on Classifying Patent Ap- plications (Molla and Seneviratne, 2018).3 The dataset and the taxonomy are presented in more detail in Section 3. Building on our previous work (Malmasi et al., 2016a; Malmasi and Zampieri, 2017), our system is based on SVM ensembles and it achieved the highest performance of the compe- tition. 2 Related Work There have been a number of studies applying NLP and Information Retrieval (IR) methods to patent applications specifically, and to legal texts in general, published in the last few years. Applications of NLP and IR to legal texts include the use of text summarization methods (Farzindar and Lapalme, 2004) to summarize le- gal documents and most recently, court ruling pre- diction. A few papers have been published on this topic, such as the one by Katz et al. (2014) which reported 70% accuracy in predicting de- cisions of the US Supreme Court, Aletras et al. (2016); Medvedeva et al. (2018) which explored computational methods to predict decisions of the European Court of Human Rights (ECRH), and (Sulea et al., 2017a,b) on predicting the decisions of the French Supreme Court. In addition to the aforementioned studies, one recent shared task …

Content truncated to 500 words