Awais Majeed

Bio: Awais Majeed is an academic researcher from Bahria University. The author has contributed to research in topics: Reputation & Emergency management. The author has an hindex of 5, co-authored 13 publications receiving 86 citations. Previous affiliations of Awais Majeed include National University of Science and Technology & University of the Sciences.

Using Crowd-Source Based Features from Social Media and Conventional Features to Predict the Movies Popularity

Abstract: Predicting the success of movies has been of interest to economists and investors (media and production houses) as well as predictive analysts. A number of attributes such as cast, genre, budget, production house, PG rating affect the popularity of a movie. Social media such as Twitter, YouTube etc. are major platforms where people can share their views about the movies. This paper describes experiments in predictive analysis using machine learning algorithms on both conventional features, collected from movies databases on Web as well as social media features (text comments on YouTube, Tweets). The results demonstrate that the sentiments harnessed from social media and other social media features can predict the success with more accuracy than that of using conventional features. We achieved best value of 77% and 61% using selected social media features for Rating and Income prediction respectively, whereas selected conventional features gave results of 76.2% and 52% respectively. More it was found that the blend of both types of attributes (conventional and those collected from social media) can outperform the existing approaches in this domain.

Towards Creation of Linguistic Resources for Bilingual Sentiment Analysis of Twitter Data

Abstract: This paper presents an approach towards bi-lingual sentiment analysis of tweets. Social networks being most advanced and popular communication medium can help in designing better government and business strategies. There are a number of studies reported that use data from social networks; however, most of them are based on English language. In this research, we have focused on sentiment analysis of bilingual dataset (English and Roman-Urdu) on topic of national interest (General Elections). Our experiments produced encouraging results with 76% of tweet’s sentiment strength classified correctly. We have also created a bi-lingual lexicon that stores the sentiment strength of English and Roman Urdu terms. Our lexicon is available at: https://sites.google. com/a/mcs.edu.pk/codteem/biling_senti

A Survey of Evolution in Predictive Models and Impacting Factors in Customer Churn

Abstract: The information-based prediction models using machine learning techniques have gained massive popularity during the last few decades. Such models have been applied in a number of domains such as me...

Vocabulary of Quranic Concepts: A semi-automatically created terminology of Holy Quran

Abstract: The identification and organization of terminology is the foremost step while organizing the domain knowledge for any domain as it is the terms and their inter-relationships that define the conceptual knowledge base. Quran, comprising the divine words of wisdom has been considered and used as prime source of knowledge and guidance for Muslims throughout the world for fourteen centuries. The concepts/topics discussed in Quran have been organized/indexed by many scholars which are used by Muslims who use them to search for guidance regarding various issues of daily life. In current era of information technology, various search services for Quranic topics are available online. They mostly use the terminologies (concepts hierarchy) manually built by scholars. In our work, we have used a semi-automatic approach to identify important concepts/topics from six English translations of Quran, and organized them into a hierarchical structure, named as Vocabulary of Quranic Concepts (VQC). CNC Value method of term recognition is used to identify significant concepts, which are then manually analyzed by domain expert, and are then organized into a hierarchy using the term-head principle. Due to extreme sensitivity of this work, complete automation of system is avoided and outcomes at all steps are manually analyzed. Currently, we have developed a vocabulary from translation of only second chapter of Quran (Al-Bakara). VQC is available at: https://sites.google.com/a/mcs.edu.pk/codteem/projects/qwn

A reputation management system for efficient selection of disaster management team

Abstract: Managing a disaster and emergency situation is a challenging task. Various ICT based systems like the Oasis and SAHANA have been developed to provide necessary collaboration, operational monitoring and resource sharing facilities for different phases of disaster management. As different organizations share their resources and skills in a disaster situation, the concepts related to collaborative networks become more relevant. Under such conditions, one of the issues is related to the efficient partner or team member selection as applicable in the case of collaborative networks. Although different partner selection mechanisms have been proposed in the literature of collaborative networks but considering the dynamic context of trust, these cannot be applied directly in the disaster management situation. Trust and reputation have been identified as one of the important factors for the efficient disaster management in the related literature. The current work focuses on the development of a reputation management system for efficient selection of disaster management team. For this, a framework identifies important factors having impact on the reputation and trust of a particular partner working in collaboration with other organizations, proposes a Service Oriented Architecture to extract information from information sources and finally proposes an algorithm for calculating the reputation score. The system can be applied in team formation and performance management system of various disaster management support tools.

A Churn Prediction Model Using Random Forest: Analysis of Machine Learning Techniques for Churn Prediction and Factor Identification in Telecom Sector

Abstract: In the telecom sector, a huge volume of data is being generated on a daily basis due to a vast client base. Decision makers and business analysts emphasized that attaining new customers is costlier than retaining the existing ones. Business analysts and customer relationship management (CRM) analyzers need to know the reasons for churn customers, as well as, behavior patterns from the existing churn customers' data. This paper proposes a churn prediction model that uses classification, as well as, clustering techniques to identify the churn customers and provides the factors behind the churning of customers in the telecom sector. Feature selection is performed by using information gain and correlation attribute ranking filter. The proposed model first classifies churn customers data using classification algorithms, in which the Random Forest (RF) algorithm performed well with 88.63% correctly classified instances. Creating effective retention policies is an essential task of the CRM to prevent churners. After classification, the proposed model segments the churning customer's data by categorizing the churn customers in groups using cosine similarity to provide group-based retention offers. This paper also identified churn factors that are essential in determining the root causes of churn. By knowing the significant churn factors from customers' data, CRM can improve productivity, recommend relevant promotions to the group of likely churn customers based on similar behavior patterns, and excessively improve marketing campaigns of the company. The proposed churn prediction model is evaluated using metrics, such as accuracy, precision, recall, f-measure, and receiving operating characteristics (ROC) area. The results reveal that our proposed churn prediction model produced better churn classification using the RF algorithm and customer profiling using k-means clustering. Furthermore, it also provides factors behind the churning of churn customers through the rules generated by using the attribute-selected classifier algorithm.

Extending the Technology Acceptance Model to Understand Students’ Use of Learning Management Systems in Saudi Higher Education

Abstract: Although learning management systems (LMS) have been widely adopted by higher educational institutions in many countries, they are considered an emerging technology in Saudi Arabia. Furthermore, research has demonstrated that the students’ use of them is not always satisfactory. This quantitative study investigated the factors that affect the students use of LMS in higher education by extending the technology acceptance model (TAM) and adapting eight external variables. Based on the probability multi-stage cluster sampling technique, online surveys were sent by email to 2000 students registered in three public universities in Saudi Arabia. 851 responses were submitted by participants, and 833 responses were used for data analysis. Using Partial Least Squares Structural Equations Modeling (PLS-SEM), the results revealed that perceived ease of use is affected by six factors (content quality, system navigation, ease of access, system interactivity, instructional assessment and system learnability). The findings confirmed that perceived usefulness has five determinants (content quality, learning support, system interactivity, instructional assessment and perceived ease of use). This research is relevant to researchers, decision makers and e-learning systems designers working to enhance students’ use of e-learning systems in higher education, in particular where there is not yet widespread adoption.

Sentiment analysis techniques in recent works

Abstract: Sentiment Analysis (SA) task is to label people's opinions as different categories such as positive and negative from a given piece of text. Another task is to decide whether a given text is subjective, expressing the writer's opinions, or objective, expressing. These tasks were performed at different levels of analysis ranging from the document level, to the sentence and phrase level. Another task is aspect extraction which originated from aspect-based sentiment analysis in phrase level. All these tasks are under the umbrella of SA. In recent years a large number of methods, techniques and enhancements have been proposed for the problem of SA in different tasks at different levels. This survey aims to categorize SA techniques in general, without focusing on specific level or task. And also to review the main research problems in recent articles presented in this field. We found that machine learning-based techniques including supervised learning, unsupervised learning and semi-supervised learning techniques, Lexicon-based techniques and hybrid techniques are the most frequent techniques used. The open problems are that recent techniques are still unable to work well in different domain; sentiment classification based on insufficient labeled data is still a challenging problem; there is lack of SA research in languages other than English; and existing techniques are still unable to deal with complex sentences that requires more than sentiment words and simple parsing.

Deep sentiments in Roman Urdu text using Recurrent Convolutional Neural Network model

Abstract: Although over 64 million people worldwide speak Urdu language and are well aware of its Roman script, limited research and efforts have been made to carry out sentiment analysis and build language resources for the Roman Urdu language. This article proposes a deep learning model to mine the emotions and attitudes of people expressed in Roman Urdu - consisting of 10,021 sentences from 566 online threads belonging to the following genres: Sports; Software; Food & Recipes; Drama; and Politics. The objectives of this research are twofold: (1) to develop a human-annotated benchmark corpus for the under-resourced Roman Urdu language for the sentiment analysis; and (2) to evaluate sentiment analysis techniques using the Rule-based, N-gram, and Recurrent Convolutional Neural Network (RCNN) models. Using Corpus, annotated by three experts to be positive, negative, and neutral with 0.557 Cohen's Kappa score, we run two sets of tests, i.e., binary classification (positive and negative) and tertiary classification (positive, negative and neutral). Finally, the results of the RCNN model are analyzed by comparing it with the outcome of the Rule-based and N-gram models. We show that the RCNN model outperforms baseline models in terms of accuracy of 0.652 for binary classification and 0.572 for tertiary classification.

Multilingual author profiling on Facebook

Abstract: Proposed a multilingual (Roman Urdu and English) author profiling corpus of Facebook profiles.Manually developed a bilingual dictionary (Roman Urdu to English) of 7749 entries and translated multilingual corpus using it.Applied 64 stylometry and 11 content based features on multilingual and translated corpora.Best results obtained using word bigram for age and word unigram, character 3 and 8 gram for gender identification. Author profiling is the identification of demographic features of an author by examining his written text. Recently, it has attracted the attention of research community due to its potential applications in forensic, security, marketing, fake profiles identification on online social networking sites, capturing sender of harassing messages etc. We need benchmark corpora to develop and evaluate techniques for author profiling. Majority of the existing corpora are for English and other European languages but not for underresourced South Asian languages, like Roman Urdu (written using English alphabets). Roman Urdu is used in daily communication by a large number of native speakers of Urdu around the world particularly in Facebook posts/comments, Twitter tweets, blogs, chat blogs and SMS messaging. The construction of sentences of Urdu while using alphabets of English transforms the language properties of the text. We aim to investigate the behavior of existing author profiling techniques for multilingual text consisting of English and Roman Urdu, concretely for gender and age identification. We here focus on author profiling on Facebook by (i) developing a multilingual (Roman Urdu and English) corpus, (ii) manually building of a bilingual dictionary for translating Roman Urdu words into English, (iii) modeling existing state-of-the-art author profiling techniques by using content based features (word and character Ngrams) and 64 different stylistic based features (11 lexical word based features, 47 lexical character based features and 6 vocabulary richness measures) for age and gender identification on multilingual and translated corpora, (iv) evaluating and comparing the behavior of above mentioned techniques on multilingual and translated corpora. Our extensive empirical evaluation shows that (i) existing author profiling techniques can be used for multilingual text (Roman Urdu + English) as well as monolingual text (corpus obtained after translating multilingual corpus using bilingual dictionary), (ii) content based methods outperform stylistic based methods for both gender and age identification task and (iii) translation of multilingual corpus to monolingual text does not improve results.