T. Raghunadha Reddy ¹, B. Vishnu Vardhan ², P. Vijayapal Reddy ³

Affiliations
1 Department of IT, Vardhaman College of Engineering, Shamshabad, Hyderabad - 500018, Telangana, IN
2 Department of CSE, JNTUH College of Engineering, Karimnagar - 505501, Telangana, IN
3 Department of CSE, Matrusri Engineering College, Hyderabad - 500059, Telangana, IN

Abstract

Author profiling is a text classification technique, which is used to predict the demographic characteristics of the authors by analyzing their written texts. Author Profiling became popular in several information technology enabled applications such as marketing, forensic analysis, psychology and entertainment. In reviews domain, most of the authors write reviews on several products without specifying their details. In this context, Author Profiling is helpful to know about the characteristics of the authors like gender, age, native language, educational background, location and personality traits by analyzing their written texts.  Most of the approaches for Author Profiling used various features like lexical features, content based features, structural features, syntactic features and semantic features to differentiate the writing style of the authors. These approaches of Author Profiling suffer from high dimensionality of features and fail to capture the relationship between the features. In this paper, a new approach is proposed to address the high dimensionality feature space problem by aggregating the term weights to find the weight of a document against the profile of the authors. The proposed approach was experimented on reviews domain to predict the gender and age group of the authors using accuracy as a measure.

Keywords

Accuracy, Age Prediction, Author Profiling, Document Weight, Gender Prediction, Pivoted Unique Term Normalization.

Full Text

   

K. C. Ravi Kumar ¹, D. Teja Santosh ², B. Vishnu Vardhan ³

Affiliations
1 Department of CSE, JNTUH, Hyderabad – 500075, Telengana, IN
2 Department of CSE, JNTUK, Kakinada – 533003, Andhra Pradesh, IN
3 Department of CSE, JNTUHCEJ, Jagtial – 505501, Telengana, IN

Abstract

Objective: A comprehensive feature extraction approach is specified by exploring the natural language rules for the extraction of various kinds of product features from Amazon online reviews. Method: The step-by-step feature extraction approach is followed to reach the goal of extracting maximum number of product features from the product reviews. Various types of nouns are extracted in the form of product features. These are namely frequent features, relevant features, implicit features and infrequent features. Findings: The results show that the comprehensive feature extraction approach performs better than the particular way for extracting the product features in the semantic environment. Applications: This approach is used in e-commerce websites to find out what product features are of interest to the customers. This model is useful in recommending products to the customers as the search for a product in the e-commerce site takes place, the features from the product reviews are helpful with the corresponding opinion orientations. This forms the basis for suggesting similar products using the calculated sentiments in the recommendation process.

Keywords

Comprehensive Product Features Extraction, E-Commerce, Natural Language Rules, Online Reviews, Opinion Targets

Full Text