World Patent Information 62 (2020) 101980 Available online 10 August 2020 0172-2190/© 2020 Elsevier Ltd. All rights reserved. Identify trademark legal case precedents - Using machine learning to enable semantic analysis of judgments Charles V. Trappey a , Amy J.C. Trappey b, * , Bo-Hung Liu b a Department of Management Science, National Chiao Tung University, Hsinchu, Taiwan b Department of Industrial Engineering and Engineering Management, National Tsing Hua University, Hsinchu, Taiwan A R T I C L E INFO Keywords: Trademark infringement Clustering Latent dirichlet allocation Precedence analysis Recommendation platform ABSTRACT Legal case precedents have a considerable impact on the development of litigation strategies. This research uses the neural network language modeling (NNLM) approach to analyze and identify judgment documents of US trademark (TM) litigation cases as precedents of a given target case. In this research, the NNLM has been trained using 4835 TM litigation documents. There are more than 800,000 words in the entire training text set including more than 150,000 vocabularies. The words in TM legal documents are vectorized to train the NN model for e- discovery of semantically correlated precedents and their features. Specifically, non-supervised machine learning (ML) methods, including clustering and Latent Dirichlet Allocation (LDA), are applied to form the TM legal document clusters, topics, and key terminologies used to characterize the TM case descriptions and precedents. The definition of the clusters, topics and corresponding key terms enhance the ability of the system to recom- mend and explain similar case judgments for any given TM case of interest or a cease and desist letter with detailed claims of infringement. Further, the intelligent approach provides macro and micro views for companies to research TM litigation trends as a means to better protect their brand equity. 1. Introduction A trademark (TM) is a legally registered sign used to represent a product, service, or organization using words, combination of words, letters, numerals, drawings, symbols, 3-D shapes, motions, and non- visible signs such as sounds, fragrances, and colors that are uniquely capable of distinguishing what is being represented [1]. Registered TMs are intangible assets that have ownership rights and may be sold, licensed, and used to develop brand equity. In the marketplace, TMs protect a company’s products and services from brand infringement, counterfeits, and the emergence of grey markets where products and services are sold without augments such as access to genuine repair parts, defect recalls, warranties, and guarantees to ensure consistent quality and ongoing customer services over the usage lifecycle. Without the protective mechanisms offered by a registered TM, customer loyalty is difficult to sustain and brand equity can easily be devalued. The value that can be accrued by a registered TM is significant and is not limited by time as long as the mark is continuously renewed, monitored, and pro- tected. TM infringing goods or services confuse consumers and are difficult to detect. Failure to detect unlawful infringement can seriously undermine the true owner’s profits and reputation. As a common business practice, companies register their brands or slogans (sometimes musical jingles or motions) as TMs to protect their commercial rights and interests to prevent potential damage linked to TM infringement [2]. Business marketing and sales activities for sustainable business ventures dependent on physical distribution channels that have evolved to include multi-channel systems integrated with Internet based e- commerce applications. Customer purchase habits have changed over the last few decades where the cost and time required for shopping has been greatly reduced with finely targeted advertisements created from online e-business records. These records facilitate greater personaliza- tion and direct communication with the most likely customers for goods and services. E-commerce enables companies to rapidly expand into international markets using focused advertising campaigns that position products and services directly against the competition or enable im- mediate first to market advantages. Increasing TM infringements dis- putes follow the trend of rapidly increasing e-commerce transactions and advertisements. For virtually integrated and multi-channel business activities, the global scope increases the difficulties of monitoring and settling infringement cases and the regulation of grey markets. In addition to complying with TM registry requirements as key intellectual * Corresponding author. E-mail addresses: trappey@faculty.nctu.edu.tw (C.V. Trappey), trappey@ie.nthu.edu.tw (A.J.C. Trappey), s106034558@m106.nthu.edu.tw (B.-H. Liu). Contents lists available at ScienceDirect World Patent Information journal homepage: http://www.elsevier.com/locate/worpatin https://doi.org/10.1016/j.wpi.2020.101980 Received 4 June 2019; Received in revised form 6 June 2020; Accepted 10 June 2020