M.Phil / MS

Permanent URI for this collectionhttps://repository.cuilahore.edu.pk/handle/123456789/36

This collection archives the complete set of theses produced by students of the COMSATS University Islamabad, Lahore Campus.

Browse

Now showing 1 - 20 of 218

Software Measurements at Small and Medium Enterprises
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2017) Muhammad Noman; FA14-MSCS-018; LHR TP 6871; Dr. Ghulam Rasool
Software measurement processes are crucial to recognize, assess, identify and measure software products and resources. However, the successful measurement processes implementation is a challenging task in small and medium enterprises (SMEs). The main goal of this thesis is to propose a measurement framework to assess the significant factors that may influence the successful measurement processes implementation in Software SMEs. A multi-method research technique is used to perform this research. A systematic mapping study is conducted to extract the significant factors, challenges, and limitations that may affect the successful implementation of measurement processes in software organizations. A search string is prepared and, applied to different databases (IEEE, ACM, Springer etc) to identify the relevant studies. In total, 24 primary studies are finalized after applying inclusion and exclusion criteria. All the primary studies are explored comprehensively and, identified success factors are categorized in seven major constructs. A measurement framework is proposed for adaptation of measurement processes in software SMEs based on the findings of the mapping study. A total of seven constructs (success factors) are defined in the proposed framework. The proposed framework is evaluated using an online survey. An online survey is conducted from the software professionals in software organizations of Pakistan. In total, 217 responses were collected and a data set is prepared. Structural equation modeling (SEM) technique is used to evaluate the data set. Statistical tools (AMOS and SPSS) are used for factor analysis and hypothesis testing. The statistical results shows that the proposed framework is acceptable due to satisfactory values of model fit indices and use of reliable items in the path diagram. In hypotheses testing, two of the seven constructs (External Forces, Management Commitment) are found to have a significant effect on successful measurement processes in software organizations. The proposed framework can be extended by adding more factors and it can also be verified in software industries of different countries.
Semantics Of Delegation In Federated Cloud
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2017) Numan Ahmad; FA15-RCS-005; LHR TP 6915; Dr. Adnan Ahmad
In federated cloud different cloud service provider allow each other to use their unused resources at some cost when their own resources are not enough. Or they collaborate with each other by delegating different services which other CSP (cloud service providers) are not providing. Before delegating they need to get pre-approval from some administrative authority which is time consuming task and can create administrative bottleneck. There should be a mechanism in federated cloud that allow the user to delegate permissions without taking pre-approval i.e. Ad-hoc Delegation. In this thesis, we explored the semantic of decentralized ad-hoc delegation in federated cloud and proposed a framework that explains the essential characteristics for delegation in federated cloud. We also developed a model, based on framework, that explore the semantics of decentralized ah-hoc delegation in federated cloud. Moreover, the model is formulated through Colored Petri Net using CPN-tools to verify it’s working.
Developing an Urdu Lemmatize
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2017) Muntaha Iqbal; FA15-RCS-010; LHR TP 6821
Lemmatization is a process of obtaining root form of a given word. Lemmatizer is an important part of Natural Language Processing (NLP) toolkit and is essential for many NLP systems e.g. Information Retrieval (IR), plagiarism and text reuse detection, Information Extraction (IE), Machine Translation (MT), Word Sense Disambiguation (WSD) etc. Urdu is a widely spoken language in the world, but very less work has been done on developing basic NLP tools for this language, one of them is Urdu lemmatizer. Since Urdu is a morphologically rich language and has words with many inflectional and derivational forms, development of an efficient lemmatizer is a challenging task and it will be useful for many Urdu NLP applications. This research work aims to develop an efficient lemmatizer for Urdu language. For this purpose, we explored two main approaches: (1) dictionary lookup approach and (2) rule based approach. Note that we also tried combinations of these two approaches. For the first approach, we build a large dictionary of 131,539 entries. In this dictionary, 3,026 are exceptional words, 22,181 are multi-word expression, and 18,267 are Named Entities (NEs) for person and 79,418 are Named Entities (NEs) for locations. Remaining 8,647 are variants of most frequent 5,000 Urdu language words. For the second approach, we extracted a total of 215 rules (67 are prefix, 128 are postfixes) from raw Urdu text. We also developed a benchmark test dataset of 1,000 Urdu words to evaluate our proposed approaches for Urdu lemmatization task. The proposed lemmatizer was evaluated using six different approaches: (1) Dictionary Lookup Approach (DLA), (2) Rule Based Approach (RBA), (3) Dictionary (DIC) + Exceptional Word List (EWL) Approach, (4) Rule Based + Exceptional Word List Approach, (5) Rule Based + Final Dictionary Approach and (6) Final Dictionary + Rule Based Approach. Evaluation was carried out using Accuracy measure. Results showed that Dictionary + Rule Based Approach gave highest Accuracy of 82.9%. We also developed a prototype system, which takes an Urdu word as input and outputs its lemma
A Methodological Exploration of BPM Community
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2017) Saman R. Abdi; FA15-RCS-007; LHR TP 6831; Dr. Abid Sohail
The study is motivated to explore the collaboration between the authors, the patterns which are being used by the authors and the areas on which the authors are working. The end product of the research is a set of generic guidelines for the field of business process management. Business process management conference, which was one of the most famous conference is been used to develop the dataset for the study. Business process management conference started since 2003 so the papers since 2003 till 2016 were collected to create the dataset. The collaboration of authors is been studied by applying certain filters over the raw data of business process management conference. The research pattern of different authors was been studied individually from each paper and later they were all combined by identifying the common or similar aspects of the processes to form a set of frameworks. The study has created a generic set of five frameworks. Each for a specific domain of business process management i.e. business process similarity, business process abstraction, business process model redesign, business process monitoring, and business process mining. The study also provides some key facts about the most collaborative authors and the domains on which the authors or the research community is mainly focusing.
Cross Genre Author Profiling Using Semantic Tagger Based Approach
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2017) Humaira Muqdes; FA15-RCS-023; LHR TP 7000; Dr. Rao Muhammad Adeel Nawab
Author profiling is a task to predict one or more traits of an author (e.g. age, gender etc.), from his/her written text. The collaborative environments have significantly increased the chance of fake author profiles, particularly over the social media. The field of automatically detecting an author’s profile from written text has potential applications in marketing, forensics, security, detecting fake profiling and harassment cases. This research work aims to explore the problem of cross genre author profiling, in which training dataset is in one genre and test dataset is in another genre. We explored three different methods for cross-genre author profiling: (1) stylometry based approach, (2) content based approach, (3) semantic tags based approach and (4) combination of previous three approaches. As far as we are aware semantic tags based approach and combination of approaches has not been previously used for cross genre author profiling problem. Evaluation was carried out using four benchmark author profiling corpora: (1) PAN-AP-14 Social Media Corpus, (2) PAN-AP-14 Hotel Reviews corpus, (3) PAN-AP-14 Blogs Corpus and (4) PAN-AP-16 Twitter Corpus. Accuracy was used an evaluation measure. Above mentioned techniques were applied for both same and cross genre problems. For same genre, best results were obtained using content based technique on PAN-AP 14-Hotel Reviews Corpus (Accuracy =55.26) for Age (Accuracy= 66.59) for gender content based techniques for PAN-AP-Blogs Corpus (Accuracy = 67.57) for age, Semantic tagger based technique (Accuracy = 78.38) for gender. Content based technique on PAN-AP-14-social media Corpus (Accuracy =55.26) for Age (Accuracy= 78.95) for gender. Content based technique on PAN-AP-16-Twitter Corpus (Accuracy =54.76) for Age (Accuracy= 74.42) for gender. For Cross Genre Training Corpus PAN-14-Blogs and testing Corpus PAN-14-Hotel reviews for age (Accuracy = 56.96) for gender training Corpus is PAN-16-Twitter (Accuracy= 58.97). Content Based Technique for Training Corpus PAN-14-Hotel reviews and testing Corpus PAN-14-Blogs for age (Accuracy = 40.81) for gender Stylistic based for training Corpus is PAN-14-Blogs (Accuracy= 59.86). Content based technique for Training Corpus PAN-14-Social Media and testing Corpus PAN-14-Blogs for age (Accuracy = 40.81) for gender Content based technique training Corpus is PAN-14-Blogs (Accuracy= 59.86). Stylistic based technique for PAN-16- Twitter training corpus and testing corpus PAN-14-Blogs (Accuracy = 40.81) for age. Content based technique for gender where testing corpus is PAN-14-Blogs (61.22) for gender.
A Structural Variants Process Models Collection for Process Similarities Evaluations
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2017) Rimsha Anam; FA15-RCS-017; LHR TP 6822; Dr. Abid Sohail
Organizations having years of involvement in business process management, regularly keep up business process model repository containing hundreds or even a large number of process models. The process models in their repositories typically evolve from different sources and are created by various stakeholders. A new process models may be by expanding or refining existing models. Therefore, process models have a tendency to collect duplicate process models which, if left unconsolidated, may develop autonomously and cause irregularities. Similarly, it frequently happens that many organizations manage the business process models that have similar objectives but involve different business unit which may differentiate one another. The similarity (Correspondence) of process models can be explored through label, structure and behavior. This study is mainly focusing structural similarity measurement. The variations in the process model though structural correspondences have already been proposed in literature. But, majority of them are conceptual i.e. not providing the implication and verification on process model collections. Moreover, the previous studies on structural changes are not yet defining any proper classification for the structural variations. The first contribution of this study is the provision process model design classification through structural variations. We have tried to explore the availability of BPMN dataset and found some existences, one of which is SAP business process repository but, that was not available publically. So, as a contribution the complete repository was created with 600 different variants of 150 original models. The original models were taken from a verified source of our research group available dataset. Furthermore, the regenerated models were once verified from domain experts and finally included. The rigor evaluations were performed on dataset both manually and automated.
Online Trust in Teen Communities
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2017) Alia Samreen; FA15-RCS-015; LHR TP 6833; Dr. Adnan Ahmad
Consistently, individuals from everywhere throughout the world connect, share and exchange data over Internet. We communicate online to share individual data, discover answers to inquiries, direct money related exchanges, play social recreations and keep up proficient and individual connections. “Online Social Network (OSN) is a set of personal contacts through which the people maintains their social identity and receive emotional support, material aid, services, information and new social contacts” [1] like Facebook, Orkut etc. Online Social Network gives a virtual situation where people can share their stories, pictures, recordings, and take part in chat rooms for online chatting with their friends and associates. In the last decade, Online Social Networks (OSN) have expanded ubiquitously. More people use OSN to share their interests and make associations, [1] likewise, OSN help people to overcome geographical impediments. Despite of its advantages OSN has also raised some privacy concerns for users. Many people have lost their jobs [86], lives [89], become bullied [4], hacked [4] etc. because of privacy issues of OSN. The situation is even worst for teens who get emotionally, socially and physically compromised due to interactions over online social networks. This is because of the fragile psychology of teens as they are more inclined to pulls of fascination [4,38], emotionally immature [3], more inclined to sensual offers [3,5] and easily grasped by fraudulent [3]. Thus, there is a need of an application that can evaluate, maintain and exhibit the community trust over acquaintances for teens. Trust is an essential part of connections in online social networks whether it is for security, deciding information access, or recommender systems [8]. In this thesis, the student has analyzed issues of teenagers and the factors that affect the formation of end user trust in online environments. After exploring the factors, the student has proposed a framework portraying various stages and their elements important to build trust among people. On the basis of framework, a reputation model is proposed, which xi ensures that a stranger or an acquaintance is trustworthy or not. Our proposed method basically judges people on the basis of their reputation, trustworthiness, honesty, benevolence, integrity, compatibility, confidence etc. The method is also implemented using dot net framework and SQL server. The student generates a new app on Facebook App. Then, the student gets Facebook App Key and App Secret Key. After that, in dot net framework the student gets built in Facebook Configuration Dll. Then, the student creates an instance of Facebook Dll Class and set its App Key and App Secret Key. Then, ask the User Current Session Token Context to be used by Facebook. Then, it will check for Permissions assigned to us by the User on FaceBook. The application was also tested and evaluated by teens on the basis of usefulness, ease of use, ease of learning, satisfaction, and system capabilities. The questionnaire was validated through inter-item and inter-variable correlation which appeared to be sufficiently significant. Moreover, Cronbach’s Alpha test was carried out for internal consistency which also appears to be statistically significant. The average users’ response over the variable and their items suggests that teens are satisfied with the proposed approach and it really helped them to identify trustworthy and well reputed friends
Colored Petri Net Based Modeling And Analysis
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2018) Usman Younas; FA16-RCS-027; Dr. Farooq Ahmad; LHR TP 5621
Efficient and safe transportation systems are essential part of modern society and developed cities. Logistics and communications play a vital role in the economy of any country. Rail transport systems are considered as backbone for freight and passenger movement across long distances. Developing a secure rail transportation system for a developing country like Pakistan is a challenging task due to weak infrastructure and less public awareness. A secure transport system can reduce chances of collision and result in saving human lives and economy. Formal modeling of a framework or a system can assist us with checking unwavering quality, precision and consistency of the framework or system. It helps us to check the validity and authenticity of the system before the actual product comes in hand. System can be viewed from different aspects through formal specification for better understanding. Modifications in the model at design level is also possible if there are any flaws in our formal model. Railway transportation systems are safety, monetary and environmentally critical systems since its failure may cause serious outcomes, for example, loss of human life, extreme wounds, and huge size of natural harms or extensive financial disasters. Rail transport is a distributed system consist of many concurrent process occurring in parallel fashion. So the safety and heterogeneous nature of this framework requires formal modeling for its development and advancement. The fundamental focus of this study is to build up a moving block railway operation system with dispatching segments to keep away trains from crashes and wrecking. Formal modeling and analysis of the operation system will also be performed to check the consistency and reliability. A core technique for the proposed research work would be the colored Petri nets (CPNs) based on discrete and concurrent system modeling, which will be used for simulation and verification of the operation model of system. CPN is the third class of Petri nets and it is a combination of Petri Nets and Standard Meta Language (SML). After model simulation, a series of analysis checks will make sure the authenticity and consistency of the model.
Toxic Comment Classification of Roman Urdu Text
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2019) Waheed Abbas; FA17-RCS-013; LHR TP 5624; Dr. Rao Muhammad Adeel Nawab
We are living in an era of technology where most of us have easy access to the inter net. Due to internet bloom, the use of social media and discussion forum, in particular for communication, increase tremendously in recent years. But this advancement also opens doors for trolls who poison these social media and forums by their abusive behav ior toward others. Detection of toxic language online is becoming an important issue in recent years. A toxic comment is defined as rude, unreasonable, or disrespectful com ments which likely to make you leave a conversation. Current methods for handling toxicity online are often heavily dependent on manual moderation, and not scalable enough to handle growing users day by day. In this thesis, we investigated the problem of toxic language detection for Roman Urdu (transliteration of Urdu in English) text because Roman Urdu is widely used to communicate online in Pakistani and Indian community. Various work has been done to detect toxicity in the English language, but due to unavailability of large Roman Urdu corpora, there is no work done so far in Roman Urdu. The aim of this thesis work is two folds: first is to develop large scale corpus for detection of toxic language in Roman Urdu, second is to apply different proven classical machine learning and state-of-the-art deep learning methods on the developed corpus. In the first step, we collected data from the YouTube comment section to develop Roman Urdu corpus. The final corpus consists of 10,863 unique comments where we have 5,887 toxic and 4,976 non-toxic comments. We performed manual annotation of corpus and got it reviewed from another annotator. We applied proven classical ma chine learning algorithms like Logistic Regression, K-Nearest Neighbors, Naive Bayes, Random Forest, Support Vector Machine and Multi-layer Perceptron along with state of-the-art deep learning methods like Long Short-Term Memory, Bidirectional Long Short-Term Memory, and Convolutional Neural Networks. Best results obtained from Random Forest with character 4-gram (Accuracy = 0.964, Precision = 0.980, Recall = 0.953, F1 = 0.966).
Detecting Urdu Semantic Textual Similarity through Word and Sentence Embedding Techniques
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2019) Muhammad Farhat Ullah; FA17-RCS-019; LHR TP 5625; Dr. Rao Muhammad Adeel Nawab
Semantic Textual Similarity (STS) evaluates the degree to which two sentences or short texts are semantically proportional to one another. STS is one of the significant prob lems in the field of Natural Language Processing (NLP). Text reuse and plagiarism detection are famous examples of STS. STS has three types; (1) Monolingual STS: if the source and suspicious short texts are in the same language. (2) Multilingual STS: if the source is in one language and sus picious short text is more than two languages. (3) Cross-lingual STS: if the source in one language and suspect is in the other language, often translation of each other. STS could be found several levels, for example, word, sentence, paragraph, and document level. Urdu is one of the low resource languages. It’s the National Language of Pakistan, also widely spoken and used in electronic, print media of Pakistan, India, and Bangladesh. The main aim of this thesis is to develop techniques that measure STS as paragraph level for the Urdu language. Our thesis aims to develop and investigate the new feature extraction techniques to ad dress the problem of STS for Urdu. We divide it into three layers, (1) Train Word and Sentence Embedding models on Urdu datasets, (2) Apply these new feature extraction techniques to extract feature from Urdu short text pairs, and (3) Apply machine learn ing classification algorithms for Urdu STS. In the first step we train Word and Sentence Embedding models on Urdu datasets. By using these embedding models, we extract the word, and sentence embedding features from pre-processed Urdu Short Text Reuse Corpus (USTRC) short texts. After that, we find the cosine similarity between these extracted feature vectors, then apply classification algorithms on similarity to classify short texts into verbatim, paraphrased and independently written. In word embedding techniques, we used Word2Vec, GloVe, and FastText with Addition, Average, and Mul tiplication Functions also we explored Smooth Inverse (SI) and Term Frequency (TF) weighted word embedding techniques. In sentence embedding techniques we have used two unsupervised (sent2vec and LASER) and two supervised (InferSent and BERT) ix techniques. We have used seven machine learning algorithms to classify similarity score including, Naive Bayes (NB), Support Vector Machine (SVM), Logistic Regression (LR), Random Forest (RF), K-Nearest Neighbour (KNN), Gradient Boost (GB), Multi-Layer Percep tron (MLP). To evaluate these classifiers, we apply F1 measure. We got best F1 measure = 0.68, 0.75, 0.92, 0.70 by using sent2vec sentence embedding technique with GB and MLP classifiers.
I-Know: Knowledge Representation and Manipulation for Self-aware Robots
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2019) Nageen Naeem; SP16-RCS-025; LHR TP 5619; Dr. Wajahat Mahmood Qazi
Knowledge representation is one of the crucial tasks while designing an artificially intelligent/self-aware agents (softbots or robots). Construction of knowledge-base and its use has been seen in past practices but with the advancement and involvement of a new generation of robotics where robot collaborate with human or with other robots. There is a requirement of knowledge-base which are not hand-crafted and the knowledge is acquired from the sensory modality interpret it and store the knowledge. The advancement in artificial intelligence (AI) design system requires explicit knowledge with the implicit knowledge. In order to extract structured knowledge from unstructured information, different information extraction techniques have been introduced. Moreover, the artificially intelligent systems are base on different cognitive architectures. These architectures at the ground level follow the same basic philosophies of cognition and regardless of school of thought the knowledge is represented using the state-of-the-art representation scheme. The utilization of these schemes are modified according to the requirement and philosophical structure of the cognitive architecture. This difference blew up the limitation of compatibility if one wants to integrate the representation scheme with other such as if representation scheme of CLARION is needed to be extended with the design rationale of atomspace they are incompatible with each other. In order to resolve this, issue this study proposed a framework for unified representation scheme which takes the best properties of the existing architecture and represents them into the atom of knowledge. The proposed representation scheme is designed to make it compatible with both softbots and robots
Periodic Mining of Users Routine Desktop Behavior for Task Assistance
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2019) SEYYAD ZISHAN ALI; FA16-RCS-028; LHR TP 5622
Now-a-days the use of technology is rapidly growing in the workplaces and a large number of people are using desktop computers for their personal and professional tasks. As a result of this growth, the large amount of data is been gathered and it’s become difficult to retrieve and manage this huge amount of data in a beneficial way. People usually face difficulty in managing documents and finding relevant files when they want to do specific task in hand. We believe that pervasive monitoring of user’s desktop activities and then mining of user’s periodic behavior can be useful in assisting the future tasks of user. A number of studies have been carried out on monitoring user desktop routine activity. However, these studies have not utilized user’s temporal routine activity patterns for assistance. It was a need to develop a desktop assistant by mining user routine behavior in accordance with the time and date more specifically weekdays and weekends. This work detects and classifies meaningful activities from large data and then models these activities into user’s one day and routine behaviors. This system reads log file and detect some meaningful activities from the large data, after identifying activities the system model these activities into user’s one day activity, then system model routine activities based on one day activity models and evaluates probability measures of user’s routine activities. After getting best results the system finds the best models of routine behavior which may use for the prediction of next possible routine activity of the user based on logged data set. It mines the best models of routine activities which may be useful for the prediction of next possible activity of the user. After a comparative results analysis of standard machine learning and data mining algorithms this research found Decision Tree best for classification of user’s activities with highest mean score of 0.92. Moreover this research has found Decision Tree as best algorithm for prediction of user’s future possible routine activities with 0.84 as highest score. The system has predicted future routine activities of a single user on Tuesday with an accuracy of 92%. This research is basically the implementation of a novel approach for mining user’s periodic desktop routine activity in accordance with the time, date and day.
Automated Detection of Early Pulmonary Nodule in Computed Tomography Images
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2019) Ahmad Usama Tariq; FA16-RCS-012; LHR TP 5777; Dr. Usama Ijaz Bajwa
Classification of lung cancer in CT scans majorly have two steps, detect all suspicious lesions also known as pulmonary nodules and calculate the malignancy. Currently, a lot of studies are about nodules detection, but some are about the evaluation of nodule malignancy. Since the presence of nodule does not unquestionably define the presence lung cancer and the morphology of nodule has a complex association with malignant growth, the diagnosis of lung cancer requests cautious examinations on each suspicious nodule and integrateed information every nodule. We propose a 3D CNN CAD system to solve this problem. The system consists of two modules a 3D CNN for nodule detec tion, which outputs all suspicious nodules for a subject and second module train on XGBoost classifier with selective data to acquire the probability of lung malignancy for the subject
Development of Large Scale English-Urdu Machine Translation Corpus for Statistical and Neural Machine Translation Systems
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2019) Moodser Hussain; FA17-RCS-005; LHR TP 5623; Dr. Rao Muhammad Adeel Nawab
The process of automatically converting the text of one natural language into an other is known as Machine Translation (MT). The two state-of-the-art machine translation techniques are Statistical Machine Translation (SMT) and Neural Ma chine Translation (NMT). In both MT techniques, a large-scale parallel aligned corpus of source & target language text is the most basic requirement to develop quality models. For the Urdu language, there are minimal parallel resources de veloped until now, and these existing resources are not enough to produce some quality translation systems. This research study is majorly focusing on: (1) the development of a large-scale semi-automatically aligned English-Urdu parallel corpus of more than 150k sen tences by collecting the data from different domains. (2) the compilation of large sized monolingual data for English and Urdu language by merging existing cor pora. (3) the development of a character-separated parallel corpus for supervised transliteration (by using Wikipedia titles) (4) the evaluation of impact generated by language model and monolingual corpus on translation quality (by using differ ent data and word-order in language models training). (5) development of Phrase based SMT (PB-SMT) and Recurrent Neural Network-based NMT (RNN-NMT) systems to evaluate and compare their performance on the proposed parallel cor pus. (6) the usage of supervised transliteration and ‘minimum error rate training’ to enhance the quality of SMT. To evaluate the performance of MT systems, ‘Bilingual Evaluation Understudy (BLEU)’ is used as an evaluation measure. The best-achieved scores for English to-Urdu translations are 37.39 and 32.70 by using PBSMT and RNN-NMT, re spectively. While for Urdu-to-English translations, PBSMT and RNN-NMT yield the best score of 28.20 and 30.63 respectively
Mono- and Cross-lingual Semantic Word Similarity for Urdu Language
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2019) Ghazeefa Fatima; FA17-RCS-016; LHR TP 5778; Dr. Muhammad Salman Khan
The process of finding a degree of similarity between words is called semantic word similarity. Machine Translation and information retrieval are common and well distinguishes applications of Semantic Word Similarity. Because of its applications, finding similarity between words is a crucial task and it becomes an important problem of Natural Language Processing. Semantic Word Similarity has two basic types: (1) Mono-lingual Semantic Word Similarity (2) Cross-lingual Semantic Word Similarity. When both words are in the same language it’s called Mono-lingual Semantic Word Similarity, on the other hand, when the both words come across two different languages it’s called Cross-lingual Semantic Word Similarity. Semantic similarity can be detected at many levels i.e. word, sentence, phrase or document level. Urdu is the national language of Pakistan. 11 million people speak it as mother tongue and 105 million people as a second language [1]. According to Riaz [2], it has been estimated that the Urdu language has 300 million speakers in the overall world. The aim of this thesis is to develop mono- and cross-lingual semantic word similarity datasets and models for the Urdu language. The purpose of this study is threefold (1) Create Monolingual Semantic Word Similarity dataset (2) Develop Cross-lingual Semantic Word Similarity dataset (3) Create new models and evaluate pre-trained models using newly created gold standard mono- and cross-lingual corpora. In the first step, we will translate SemEval-500 dataset into Urdu language using Google, Bing and other translation resources. After translation, a linguistic expert verifies the translation and make corrections as required. The verified dataset then assigned to 12 annotators for annotation by following the guidelines provided in SemEval-2014 task 3 [3]. In the second step, the newly created dataset and X SemEval-500 dataset are combined to create a cross-lingual dataset. This dataset is also annotated by 12 annotators by following the same guidelines. We use intra-class correlation coefficient for inter annotator agreement of both mono- and cross-lingual dataset and got r=0.9 with p < .001 for each dataset. In the third step, we create new models for Monolingual semantic Word Similarity and Cross-lingual Semantic Word Similarity. We evaluate these newly created models and pre-trained models using our mono and cross-lingual gold standard datasets. We use person correlation as evaluation measure. In case of monolingual models, we got correlation of 0.54 for corpus based techniques. Fasttext pre-trained Cbow and Skip-gram models got correlation of 0.53 and 0.63 respectively. Ur-Mono based trained fasttext Cbow and Skip-gram models got maximum correlation of 0.63 and 0.67 respectively. For cross-lingual semantic word similarity, we applied Translation Plus Monoligual analysis technique, in which after translation of word pairs, multiple WordNet measures are applied. Path Similarity measure outperform other measures with correlation of 0.70 while using Google translation. In case of Bing translation, Path Similarity and Leacock measure outperform other measures with correlation of 0.71 and 0.71 respectively
Urdu-English Machine Transliteration using Neural Networks
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2019) Usman Mohy ud Din; FA16-RCS-007; LHR TP 5776; Dr. Muhammad Waqas Anwar
Machine translation has gained much attention in recent years. It is a sub-field of computational linguistic which focus on translating text from one language to other language. Among different translation techniques, neural network currently leading the domain with its capabilities of providing a single large neural network with attention mechanism, sequence-to-sequence and long-short term modelling. Despite significant progress in domain of machine translation, translation of out-of-vocabulary words(OOV) which include technical terms, named-entities, foreign words are still a challenge for current state-of-art translation systems, and this situation becomes even worse while translating between low resource languages or languages having different structures. Due to morphological richness of a language, a word may have different meninges in different context. In such scenarios, translation of word is not only enough in order provide the correct/quality translation. Transliteration is a way to consider the context of word/sentence during translation. For low resource language like Urdu, it is very difficult to have/find parallel corpus for transliteration which is large enough to train the system. In this work, we presented transliteration technique based on Expectation Maximization (EM) which is un-supervised and language independent. Systems learns the pattern and out-of-vocabulary (OOV) words from parallel corpus and there is no need to train it on transliteration corpus explicitly. This approach is tested on three models of statistical machine translation (SMT) which include phrase based, hierarchical phrase-based and factor based models and two models of neural machine translation which include LSTM and transformer model. On SMT models, there is gain of 0.63 to 0.91 in BLEU score while on NMT models, there is gain of 1.28 to 2.05 in BLEU which are better than previous baseline scores. Our approach shows promising results in translation of Urdu text into English which is mostly neglected due to its complexities. We also discussed the results, different challenges faced during this work and effect of right pre-processing techniques
Multi-label Document Classification for Urdu Language
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2019) Madiha Rasool; FA17-RCS-010; LHR TP 5779; Dr. Muhammad Salman Khan
Multi-label document classification is a conventional supervised machine learning task where each document is associated with multiple labels to correctly categorize the meaning of context. All natural languages present this computational linguistic task for a wide series of research domains, such as Natural Language processing, document categorization and text mining. Many benchmark corpora have been developed for multi-label classification task, but most are for English and other European languages. However, there is a dearth of annotated benchmark corpus and methods for the multi-label document classification task for the Urdu language. To fill this gap, we proposed a large benchmark corpus and methods for the multi-label classification of Urdu documents. The proposed corpus contains 600 documents from the field of journalism in the following domains: sports, national, foreign, showbiz and business. The proposed corpus has been manually annotated with USAS (UCREL Semantic Analysis System) scheme which provides 21 major labels and 232 sub labels for fine grained annotation. Each document in the proposed dataset was annotated with minimum two labels and maximum six labels to present a detailed analysis of the context as a supervised multi-label document classification task. To demonstrate how our proposed dataset can be used for the development and evaluation of Urdu multi-label document classification system, three different techniques were applied: (1) Multi-Word Expression (MWE) N-gram (2) Single-Word Expression (SWE) N-gram and (3) Most Frequent Word (MFW) N-gram. Seven different multi-label classifiers were applied on features extracted using three MWE, SWE and MFW methods. Evaluation was carried out using accuracy, exact match, hamming loss and F1 measures. Results show that the (F1 is 0.545) gives best results by using MFW technique on CC and MCC classifier
Urdu Word Sense Disambiguation using Siamese Neural Networks
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2020) Waqar Ashiq; SP19-RCS-013; LHR TP 6419; Dr. Rao Muhammad Adeel Nawab
The process of Word Sense Disambiguation (WSD) is to recognize the exact sense of the targeted polysemous terms in the given context. Word sense ambiguities exist in all human languages, and humans can easily solve this ambiguity but it’s a tough task for machines to solve automatically. WSD is a significant long-established task in the area of natural language processing (NLP) with a remarkable contribution in numerous applications. The task of WSD has been categorized into (1) Lexical Sample, and (2) All-Words WSD. In literature, the problem of Urdu WSD has been mainly addressed using feature-based machine learning methods (based on manual feature engineering). A major limitation of manual feature engineering is that it requires a lot of time and endeavors to develop machine learning methods in Urdu WSD. To overcome the limitations of manual feature engineering, deep learning approaches are proposed, and they have produced promising results for various NLP tasks. To fulfill the gap in WSD research, we developed and apply state-of-the-art Siamese Neural Network (SNN) based approaches for both tasks for the Urdu Language. We developed the SNN based approaches for both WSD tasks using six deep learning models (CNN, RNN, LSTM, BiLSTM, GRU, and BiGRU) with the combination of three distance measure functions including Euclidean distance, Manhattan distance, and Cosine similarity. These Siamese network features of all deep learning models are combined for WSD classification using eight machine learning classifiers with two well-known feature selection methods (RFECV and Chi-Square). Additionally, for the All-words task SNN features are combined with n-gram (unigram and bigram) features. The developed SNN based approaches are evaluated on well-known benchmark corpora which include (1) UAW-WSD-18 Corpus for All-words WSD and (2) ULS WSD-18-Corpus for Lexical sample WSD task. After extensive experimentation, the proposed approach produced state-of-the-art results (Accuracy = 64%) in All-words WSD and outperform the baseline results (Accuracy = 57.71%). The results of Lexical sample WSD are not improved because of very few instances of each ambiguous word in corpora for deep learning approaches.
Multi-Label Author Profiling on Multi-Lingual Text
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2020) Samra Kanwal; SP19-RCS-011; LHR TP 6418; Dr. Rao Muhammad Adeel Nawab
Author profiling is the task of author attributes classification where the main aim is to predict the profile and demographic features of an author which includes age group, gender, region, personality, etc., by examining the written content of the author. There are different promising applications of author profiling including security, forensic analysis, and identification of harassing text messages, marketing intelligence, and fake profile identification. In literature, the majority of the studies have been carried for single-label author profiling i.e., predicting only one single label at a time. There are very few studies available on multi-label author profiling on mono-lingual text, i.e., predicting more than one label at a time. However, the problem of multi-label author profiling has not been completely explored for multi-lingual text. The main objective of this research work is to explore the problem of multi-label author profiling on multi lingual text (English and Roman Urdu). For this purpose, the aim is to predict four author traits including gender, age, education, and language as a multi-label task using three state-of-the-art methods: (1) Content based Methods (N-gram models for both word and character), (2) Deep Learning Approaches (CNN, LSTM, BI-LSTM, GRU, and BI-GRU) and (3) Transfer Learning Approaches (BERT, and XLNET). The evaluations were carried out on three benchmark multi-lingual datasets, RUEN-AP-17, SMS–AP–18, and BT-AP-19. After extensive experimentation and comparison, the results show that the content-based method outperforms the deep learning and transfer learning methods for multi-label author profiling tasks on all multi-lingual corpora used in this study. On the RUEN AP-17 corpus the best results (Accuracy = 0.71, F1-measure = 0.65) were obtained using the word tri-gram model with the Naïve Bayes classifier. On SMS–AP–18 corpus the best results (Accuracy = 0.74, F1-measure = 0.69) were obtained using word uni gram model using support vector machine with one-vs-rest and one-vs-one classifiers, and on BT-AP-19 corpus the best results (Accuracy = 0.74, F1-measure = 0.69) were obtained using word bi-gram model using support vector machine with one-vs-rest and one-vs-one classifiers
A Game Theory Based Trust Model for Internet of Things
(Library Information Services, COMSATS University Islamabad, Lahore Campus, 2020) Ghulam Mustafa; FA15-RCS-029; LHR TP 5970; Dr. Adnan Ahmed
Internet of Things is a paradigm shift in the advancement of wireless sensor networks. This has been realized with advanced techniques of service provision and exchange of data. The concept behind this technology is “anytime, anywhere, any media”. There is a huge potential in IoT for the development of a large number of application out of which very small part has been deployed in our lives. Devices connected through the internet of Things are equipped with smart objects and these objects are increasingly becoming part of our life. IoT devices are providing services in almost all walks of life like services in automated industry, medical field, chemical plants, inventory control and other areas of human life. IoT devices are light weight with low power and low memory. Therefore, these devices are resource constrained and data possessed by these devices is very sensitive. In this research work the author identified a number of trust factors for the differentiation of the trustworthy and suspicious devices in IoT environments. The suspicious devices some time behave abnormally due to number of facts like memory and battery constraints and low processing power. Due to their malfunctioning these devices compromise the overall trust profile of the IoT network. The author has proposed a framework through which he described the essential elements for the identification of malicious nodes. This framework helps in in identifying the suspicious behavior of giving false feedback as badmouthing against the trustworthy nodes. After framing the proposed frame work, a model has been designed in which game theory has been incorporated for the behavior correction of the malicious devices through carrot and stick policy. On the basis of the game strategy the Nash equilibrium has been achieved so that a device after choosing its strategy cannot deviate from its strategy unilaterally. The stated model has been formalized in Microsoft Azure IoT Central, a software as a service (SaaS) based cloud environment and simulation has been performed. In the simulation, the deployment of the IoT devices and their feedback values regarding the tasks and services have also been collected for the identification of badmouthing in these feedbacks. The trust value of the participating nodes has been calculated on the basis of the feedback values after passing through the proposed non cooperative game so that our proposed trust model become trustworthy and cooperation enforcing. Finally, a conclusion regarding the performance of proposed work has been drawn