International Journal of Computer Science Trends and Technology (IJCST) – Volume 4 Issue 5, Sep - Oct 2016 ISSN: 2347-8578 www.ijcstjournal.org Page 147 A Model for Employing Semantic Role Labeling To Extract Predicate Argument Structure Hadia Abbas Mohammed Elsied [1] , Naomie Salim [2] , Atif Khan [3] Department of Computer Science [1] , Sudan University of Science & Technology Khartoum, Sudan Faculty of Computing [2] , Universiti Teknologi Malaysia, 81310, Skudai, Johor, Malaysia Faculty of Computing [3] , Universiti Teknologi Malaysia, 81310, Skudai, Johor, Malaysia ABSTRACT In this study we aim to represent a dataset from sentence level form to predicate argument structure form, which is considered as a higher- level of abstraction. This new representation can be processed further in various applications such as text summarization and plagiarism detection .We use SRL (Semantic Role labeling) to identify sentence constituents then we implement a model to extract the predicate argument structure from the sentences that undergo SRL automatically , we compare our results with a manual predicate argument structure extraction, we got a good result according to precision and recall values. Keywords :— SRL , NLP , predicate Argument Structure I. INTRODUCTION Semantic Role Labeling (SRL) has been widely applied in text content analysis tasks such as text retrieval[1], information extraction[2], text categorization [3] and sentiment analysis [4]. In the area of text summarization, [5]introduced a work that combined semantic role labeling with general statistic method (GSM) to determine important sentences for single document extractive summary, also [6] introduce a work of abstractive summarization uses SRL. The sentence-level semantic analysis of text is related with the characterization of events, such as figuring out “who” did “what” to “whom,” “where,” “when,” and “how.” “what” took place is established by the predicate of the statement or the clause , and the other remaining sentence constituents expresses the contributors in the event (such as “who” and “where”), as well as further event properties (such as “when” and “how”).The main task of semantic role labeling (SRL) is to indicate exactly what semantic relations hold among a predicate and its associated participants and properties. Exemplary roles used in SRL are labels such as Agent, Patient, and Location for the entities participating in an event, and Temporal and Manner for the characterization of other aspects of the event or participant relations. This type of role labeling thus yields a first- level semantic representation of the text that indicates the basic event properties and relations among relevant entities that are expressed in the sentence [7]. II. SEMANTIC ROLE LABELLING (SRL) SRL is a task in natural language processing (NLP ) consisting of detection of the semantic arguments associated with the predicate or verb of a sentence and their classification to their specific roles , more over it is the underlying relationship that a participant has with the main verb in the clause [4], also known as semantic case, thematic role, theta role (generative grammar), and deep case (case grammar). The goal of SRL is to discover the predicate argument structure of each predicate in a given input sentence[5] . According to [6] the task of SRL is to find all arguments for a given predicate in a sentence and label them with semantic roles. Semantic role labeling (SRL) is a process to identify and label arguments in a text. SRL can be extended for the events characterization task that answer simple questions such as “who” did “what” to “whom”, “where”, “when”, and “how”.The main task of SRL is to show what specific relations hold among a predicate with respect to its associated participants . As the definition of the PropBank and CoNLL- 2004 shared task [10] there are six different types of arguments labeled as A0-A5 and AA. These labels have different semantics for each verb as specified in the PropBank Frame files. In addition, there are also 13 types of adjuncts labeled as AM-adj where adj specifies the adjunct type. SRL aims to identify the constituents of a sentence, with their roles such as Agent, Patient, Instrument etc., and the adjunctive arguments of the predicate such as Locative, Temporal, with respect to the sentence predicates [3]. This type of role labeling thus produce a first level semantic representation of RESEARCH ARTICLE OPEN ACCESS