678 The International Arab Journal of Information Technology, Vol. 18, No. 5, September 2021 Headnote Prediction Using Machine Learning Sarmad Mahar 1 , Sahar Zafar 2 , and Kamran Nishat 1 1 CoCIS, PAF-Karachi Institute of Economics and Technology, Pakistan 2 Computer Science, Sindh Madressatul Islam University, Pakistan Abstract: Headnotes are the precise explanation and summary of legal points in an issued judgment. Law journals hire experienced lawyers to write these headnotes. These headnotes help the reader quickly determine the issue discussed in the case. Headnotes comprise two parts. The first part comprises the topic discussed in the judgment, and the second part contains a summary of that judgment. In this thesis, we design, develop and evaluate headnote prediction using machine learning, without involving human involvement. We divided this task into a two steps process. In the first step, we predict law points used in the judgment by using text classification algorithms. The second step generates a summary of the judgment using text summarization techniques. To achieve this task, we created a Databank by extracting data from different law sources in Pakistan. We labelled training data generated based on Pakistan law websites. We tested different feature extraction methods on judiciary data to improve our system. Using these feature extraction methods, we developed a dictionary of terminology for ease of reference and utility. Our approach achieves 65% accuracy by using Linear Support Vector Classification with tri- gram and without stemmer. Using active learning our system can continuously improve the accuracy with the increased labelled examples provided by the users of the system. Keywords: Judgment summary, head-note prediction, machine learning, text summarization. Received March 6, 2020; accepted September 17, 2020 https://doi.org/10.34028/iajit/18/5/7 1. Introduction Currently, Legal domain analysis has become an attractive research field for researchers. More and more intelligent systems are explored using legal documents, including Judgments/Orders Summarization, topic prediction using machine learning and Information retrieval [22]. In this research, we design, develop and evaluate headnote prediction using machine learning, without involving human interference using Supervised Learning algorithms and predict head-notes [5]. Head-notes are an interpretation of Judgment/Order written by the editor. Mostly legal Judgments are complex, lengthy and refer to other Judgments. In addition, it is citing as a ‘precedent’ in a given set of circumstances. These judgments later referred to in other cases [12, 20]. Headnotes divided into two main parts; i.e., one Law point that discussed in Judgment/ Orders and another is the editor’s interpretation of that law into Judgment. However, they are only the editor’s remarks and not the Courts’ [7]. We summarize our key contribution as follows: 1. We have developed a data set for the experiment on judicial data to generate head-notes; there are few data sets available on different websites, however, these data sets are not relevant to the British Legal system. 2. We have developed the first-ever system in Pakistan for public sector organization free of cost with the aim that the public at large and legal professionals take advantage of this system and also encourage legal research. Active Learning Model designed a system that improves itself continuously if the user(s) update any part of the head-note or label either they accept predicted law points or reject it; we keep a record of these changes. This completely separate system performs the job of active learning and updating trained models. We design, develop and evaluate headnote prediction using machine learning, without involving human interference. 2. Background All judgments that have been accepted for publication, along with headnotes, are published in law digests/journals. Some publishers also provide a law- site (website) with various search requirements for the public and litigants' convenience [10]. The publishers hire lawyers to review the Judgments, write the gist of the Judgments’ and head- notes. Senior lawyers for proofreading prior publication review these judgments [6]. The editor must convey the highlighted reasoning of the judge and observation of the judge into the judgment [16]. Our research project will be concentrated mainly on proposing a unified organization that writes headnote automatically using a system’s trained model that will reduce the human effort involved in writing headnote and a great deal of manpower involved in writing headnote / legal editorial [2].