Overview of the 2018 ALTA Shared Task: Classifying Patent Applications

Overview of the 2018 ALTA Shared Task: Classifying Patent Applications Diego Moll´a Department of Computing Macquarie University Sydney, Australia diego.molla-aliod@mq.edu.au Dilesha Seneviratne Queensland University of Technology (QUT) Brisbane, Australia d.dwmhakmanawalawwe@hdr.qut.edu.au Abstract We present an overview of the 2018 ALTA shared task. This is the 9th of the se- ries of shared tasks organised by ALTA since 2010. The task was to classify Aus- tralian patent classifications following the sections defined by the International Pa- tient Classification (IPC), using data made available by IP Australia. We introduce the task, describe the data and present the results of the participating teams. Some of the participating teams outperformed state of the art. 1 Introduction When a patent application is submitted there is a process where the application is classified by ex- aminers of patent offices or other people. Patent classifications make it feasible to search quickly for documents about earlier disclosures similar to or related to the invention for which a patent is applied for, and to track technological trends in patent applications. The International Patent Classification (IPC) is a hierarchical patent classification system that has been agreed internationally. The first edition of the classification was established by the World In- tellectual Property Organization (WIPO) and was in force from September 1, 1968 (WIPO, 2018). The classification has undertaken a number of re- visions since then. Under the current version, a patent can have several classification symbols but there is one which is the primary one. This is what is called the primary IPC mark. An IPC classification symbol is specified ac- cording to a hierarchy of information. The generic form of the symbol is A01B 1/00, where each component has a special meaning as defined by WIPO (2018). The first character of the IPC clas- Symbol Section A Human necessities B Performing operations, transporting C Chemistry, metallurgy D Textiles, paper E Fixed constructions F Mechanical engineering, lighting, heating, weapons, blasting G Physics H Electricity Table 1: Sections of the IPC sification symbol denotes the first level of the hi- erarchy or section symbol. This is a letter from A to H as defined in Table 1. The goal of the 2018 ALTA Shared Task is to automatically classify Australian patents into one of the IPC sections A to H. Section 2 introduces the ALTA shared tasks. Section 3 presents some related work. Section 4 describes the data. Sec- tion 5 describes the evaluation criteria. Section 6 presents the results, and Section 7 concludes this paper. 2 The 2018 ALTA Shared Task The 2018 ALTA Shared Task is the 9th of the shared tasks organised by the Australasian Lan- guage Technology Association (ALTA). Like the previous ALTA shared tasks, it is targeted at uni- versity students with programming experience, but it is also open to graduates and professionals. The general objective of these shared tasks is to intro- duce interested people to the sort of problems that are the subject of active research in a field of nat- ural language processing. There are …

Content truncated to 500 words