Browse
Recent Submissions
Item iAcetylK-PseAAC: Improving Accuracy of Lysine Acetylation Predictor by Incorporating Statistical Moments and Position Relative Features into PseAAC(UMT, Lahore, 2019) AQSA ANWARAmong different Post-Translational Modification (PTM) the most vital one is lysine acetylation in protein. Its importance can’t be undermined related to different diseases and essential biological practice. The key step to find the underneath layer of acetylation along with their site is to completely apprehend the mechanism behind this biological process. In the previous research models based on artificial neural network (ANN), they have used different techniques like position weighted matrix (PWM), support vector machine (SVM), k nearest neighbors (KNN) score and many others but still they weren’t able to maximize the accuracy of the prediction. Our predictor model have used SVV, SM, FV, PRIM, RPRIM, AAPIV and RAAPIV based on ANN to compute the accuracy, sensitivity, specificity and MCC which are 82.9918%, 98.2242%, 94.504% and 0.50226 respectively using 10-fold cross validation. The results of independent dataset testing were 87.74% accuracy, 95.72% sensitivity, 70.19% specificity and 0.7067 MCC. Our model have given the more accuracy than other research models, using ANNItem U-NET: A NOVEL METHOD FOR LUNG SEGMENTATION OF CHEST WITH CONVOLUTIONAL NEURAL NETWORK(UMT, Lahore, 2019) MUHAMMAD AWAIS MALIKTuberculosis is a major health threat in many regions of the world. Opportunistic infections in immunocompromised HIV/AIDS patients and multi-drug-resistant bacterial strains have exacerbated the problem, while diagnosing tuberculosis still remains a challenge. Medical images have made a great impact on medicine, diagnosis, and treatment. The most important part of image processing is image segmentation. In this thesis I have described the latest segmentation methods applied in medical image analysis, I propose a novel method of X-ray of lungs segmentation using U-Net model. I construct the U-net which combine the lungs and mask. Then I convert to problem of positive and negative TB lungs into the segmentation of lungs, and extract the lungs by subtracting the chest from the radiography. In experiment, our model achieve 97.62% on the public dataset of collection of by Shenzhen Hospital, China and Montgomery County X-ray Set.Item IDENTIFICATION OF SEQUENCE BASE PROTO ONCOGENES BY INTEGRATION OF STATISTICAL MOMENTS INTO PseAAC(UMT, Lahore, 2019) MUHAMMAD FAROOQProto-oncogenes are a group of genes that cause normal cells to become cancerous when they are mutated. Proto-oncogenes encode proteins that function to stimulate cell division, inhibit cell differentiation, and prevent cell death. While the prediction of the proto-oncogene may happen at different phases of the cancer-causing processes, the method of prediction is always a question. Prediction through in vitro experimentations is considered sometimes a standard procedure, but is very time taking, laborious and costly. This problem can be address by opting computer aided approaches i.e. bioinformatics and computational biology. Keeping this in mind, an effective new method is proposed in this study for the prediction of proto-oncogenes. The predictor proposed in this study calculates statistical moments and position-based features and incorporates them in PseAAC by using the Chou’s 5-step rules. Later on, Random Forest is used as classifier for the accurate prediction of results. The method was validated using the 10-Fold cross-validation, Jackknife testing, Self-Consistency and Independent testing, giving 95.44%, 95.21%, 97.38%, and 96.41% accurate results, respectively. These results depict that the proposed model can play a key role in the prediction of proto-oncogenes to aid the scientists in discovery of mechanism against cancer.Item iLytic-PseAAC: Prediction of Cell wall lytic and non-lytic enzymes with highest accuracy of the predictor by integrating statistical moments and position relative features into Chou’s PseAAC(UMT, Lahore, 2019) ANAM SHAHZADIBacteriophage producing lytic enzymes are anti-infective molecules that are capable of digesting cell wall of targeted bacteria, specifically of gram-positive bacterium, causing effective bacterial cell wall lysis and consequences in ultimate death of the targeted bacterium. This lysin driven phenomenon of bacterial death is an operative tool to operate against antibiotic resistance of pathogenic bacteria due to their specificities for the pathogen and low bacterial resistance towards lysin. Thus, for controlling antibiotic resistance and infection diseases, the discrimination of lytic from non-lytic enzymes is somewhat extremely crucial that requires the reliable and comprehensive computational method that can precisely predict and discriminate the two groups of lysins. This study comprehends the construction of novel prediction model to serve the proposed purpose. We developed the prediction model based on artificial neural network by integrating the position relative features and sequence statistical moments in PseAAC for training neural networks. Highest overall accuracy has been achieved through cross-validation with the Jackknife testing that was computed to be 93.60%,93.06 % sensitivity, and 94.06 % specificity. Our astonishing experimental results demonstrated that the proposed predictor surpass the existing models that can be served as a time and cost-effective stratagem for designing novel drugs to strike the contemporary bacterial infection.Item AHP SWOT ANALYSIS FRAMEWORK PROPOSED FOR CRICKET PLAYERS PERFORMANCE ANALYSIS(UMT, Lahore, 2019) Zaman AshrafCricket is one of the most popular game in the world and competition is increasing among the teams. Every cricket board is putting its resources to search the best players and train them to overcome their opponents. For this, they use different tools and techniques to judge the abilities and weakness of the players. We are presenting a tool known as SWOT analysis through Analytic Hierarchy Process (AHP) that can help players to grow stronger and find potential opportunities. SWOT (Strengths, Weaknesses, Opportunities and Threats) analysis is a widely used technique for analyzing external and internal components and find out an efficient method to support for the decisions. The proposed technique is acquired by performing Analytic Hierarchy Process (AHP). In conclusion, the SWOT and AHP integration may provide great assistance to cricket board in determining the best players that play an important role in winning the cricket match.Item Learning based approach based on LEAR for facial fiducial point prediction of non-fronatal faces(UMT, Lahore, 2019) SANA BASHARATFacial features points (FFP) are used for localization and representation of the prominent features of the face that are traced on tips, corners and mid-points of facial segments. The Proposed algorithm to detect facial landmarks on frontal and non-frontal face on video sequences. It syndicates regression based approach with advance block matching technique. In proposed methodology,it will be shown that how tree used to take estimate position of facial landmarks, straight from a scant subsection of pixel intensities with less time and by handling missing labeled data especially for non-frontal images. The Proposed algorithm, encompass the regression based model to provide a quality measure of each prediction and use the shape model to restrict and correct the sampling region. Our approach is extension of existing LEAR combines the low computational cost with selection of important features based on 22 facial points. The proposed algorithm is tested on five datasets. Results presented significant enhancement over the current state of the art.Item CONDUCTING USER RESEARCH FOR DESIGNING BETTER SOFTWARE PRODUCTS(UMT, Lahore, 2019) MAHAM IRFANA user’s experience with a software product is crucial for the success of that product. For a long period of time, it was not considered to get an insight of what the users demanded from a product. User research is the best way to discover what the users need. With the increased use of mobile devices, the demand of software products has increased. Building an application is not as simple as hiring someone to code and deploy the application. A process has to be followed and user’s requirements must be understood before we start the design process. The research will help to take additional benefits from the results of user research. It would help to investigate and validate product ideas clearly. As more facts can help to reach a better conclusion, the results from multiple sources like interviews, surveys, user testing etc. Surveys has been conducted for different application i.e. E-commerce application and Facebook application. Proposed user experience (UX) model has been implemented on results and understands that how user experience (UX) research model helps to extract useful information. The recommended model is helpful in patching all the problems that come along during the user research process. The results produced by experiments can be processed through Atomic user experience (UX) model and a useful, shareable and easy to understand outcome can be drawn out of the research.Item ACL TEAR DIAGNOSIS FROM KNEE MRI USING CNN AND RNN ARCHITECTURE(UMT, Lahore, 2019) OMER SAIFAccurate diagnosis always has been the basis of the best suitable treatment for the patient. As far as knee ligament injuries are concerned MRI is utmost preferred method for identification of exact abnormality. However, radiologists working on interpretations is time consuming, tiring and may be wedded to errors. By using computer aided diagnosis such errors can be omitted quite efficiently and will also serve its purpose of correctly identifying the pathology. Deep learning method has been introduced in which layers of features have been embedded which is suitable for the mingled relations that exist between the images and the results. DL methodologies have outperformed conventional methods of image analysis and empowered huge advancement in therapeutic imaging assignments. Following study has been conducted to detect ACL tear by using CNN-RNN method in knee MRI. This will make its mark in identifying the patients at risk and serves its best in clinical decision making. By automatizing the process and by using CNN techniques for feature extraction and LSTM for unique decision support for detection of tears. ACL tears can be detected either partial or complete. Thus, increasing specificity and sensitivity of MRI in knee related ligamentous injuries.Item SMART CAMPUS THE FUTURE UNIVERSITY CAMPUS BASED ON IoT(UMT, Lahore, 2019) TREZAH IFTIKHARThe huge revolution in the IT industry has impacted our surroundings greatly. Our environments are now smarter than ever. Security monitoring has also been revolutionized through use of Global Positioning System (GPS) and other advanced monitoring technologies including Google Maps API and high end Location sensors. GPS represent an embedded technology that is used in smart phones to gather information related to location of different objects in our surroundings. The information is then analyzed and the results which are of great importance are also backed up at the cloud server, are then communicated to target users and used through the smart devices using smart applications. The devices that are IP enabled and are interconnected through the common connection point “the web” as a whole, called IoT or the internet of things. The proposed research indicates the use of Google Maps API in combination with IoT to design a smart campus model that is used to communicate and share location information among its users, and is secure. The shared location information is used by the proposed model for an automated attendance management system. The smart campus model is implemented and tested using android integrated development environment (IDE). The proposed smart campus environment is beneficial in terms of information accessibility and secure communication in campusItem A POLICY RECOMMENDATION FRAMEWORK TO RESOLVE GLOBAL SOFTWARE DEVELOPMENT ISSUES(UMT, Lahore, 2019) MUJTABA HASSANGlobal software development (GSD) is basically a development which is done through low cost in given time frame by sitting in remote areas within cities, countries and around the globe. The global software development is facing major challenges such as, time zone differences, language barrier, cultural differences, geographical distance, communication gap, coordination problems, lack of project guidance, lack of experienced software testers, customer dissatisfaction, poor project management and so on. These challenges has major impact on software quality in GSD and decreases the acceptability among its users. In recent years, some strategies had been applied to reduce quality related issues that exist in GSD, however there is still a need to develop effective and efficient methods, techniques and best practices that can lead to enhance the quality of software development. The main objective of this study is to highlight existing issues of GSD and to provide the policy recommendations to mitigate them. Therefore, a quantitative analysis have been performed on collected data to configure the issues of global software development that directly or indirectly effects the quality of software product. It also provides detailed statistical analysis on data by analyzing and applying demographic profile, frequencies, descriptive studies, reliability test (achieving significance (0.72 > 0.70)), nonparametric test (rejected the null hypothesis with less than 0.5 significance) and one sample T-test (with 2-tailed significance difference 0.00) through SPSS. Hence, this work contributes by making strong policy recommendations to resolve global software development issues. These policy recommendations explain how GSD project should be handled by selecting best process, and how to achieve maximum quality of software product for customers to satisfy them, improve the acceptance among its users, bring their trust back to gain more business, and overall development in (software) industry.Item Data Preservation and Digital Forensics for Virtual Machines(UMT, Lahore, 2019) Taimour NazarCloud computing is an emerging trend these days. It offers computation and storage at a relatively low-cost due to its pay per use policy. However, it has created new concerns regarding security, as all the conventional methodologies and tools for investigation fall short for cloud computing investigation. Study of recent research papers has shown that no definite strategy exists to cater this issue. Certain methodologies have been proposed by researchers but a major issue, i.e., loss of records which is vital for digital forensics due to termination of virtual machines, remains unsolved and unaddressed. Our main aim is to address this issue and propose a possible and practical solution for it. All terminating virtual machines cannot be stored for forensic because of high cost of storing huge amount of data. In our solution only relevant data of virtual machines will be stored as an XML file. Further a list of software is extracted from this XML file and it is used to find out how much risky is this virtual machine and it can give an idea to forensics experts that what type of malicious activity could have been conducted with it before it was terminated.Item MINING THE TELECOMMUNICATION DATA FOR FRAUD DETECTION ON CALL DETAIL RECORD(UMT, Lahore, 2019) JAHANZEB NAEEMThe telecommunication industry is growing rapidly and telecom companies are collecting and recording huge amounts of data that can be used to extract interesting patterns. Different data mining methods can be applied and based on the knowledge obtained, we can set up new strategies for tasks like fraud prevention, promotions, offering bundle packages for prepaid and post-paid customers, and many others. Telecom companies in Pakistan provide a variety of services to win over their subscribers because of the tough market competition. Fraud is a multi-billion-dollar problem around the globe. Telecommunication fraud causes a huge loss of income and it can affect the performance and credibility of telecommunication companies. The most difficult problem that the industry faces is that fraud is dynamic. Over the years, fraud has become a major issue to the extent that losses to telephone companies are measured in terms of millions and billions of American dollars. Fraud adversely impacts the telephone company in four ways shareholder perceptions, customer relations, financially and marketing. In this work, our main focus is on analysis of outgoing calls data to identify potential fraud cases. We further look at the vulnerability of the system and predict if the same type of fraud can occur in future. Our proposed approaches are based on data mining and machine learning techniques including SVM, Artificial Neural Networks and Naïve Bayes. We achieve an accuracy of 93.19% for vulnerability prediction and for fraud detection we achieve an accuracy of 93.2%.Item SOFTWARE PROJECT MANAGEMENT EDUCATION: A SYSTEMATIC LITERATURE REVIEW(UMT, Lahore, 2019) IFAT UN NISAHSoftware project management (SPM) is the emerging field of software engineering which attracted the huge number of researchers and developers in recent 5 to 10 years. Nowadays, Software project management education (SPME) is playing significant role to create professionals which has competency to complete and execute the software projects successfully. As the demand of Software project management (SPM) is exponentially increasing, the responsibilities of academia increased to ensure and provide the solid base to software engineering students. There are different distinctive methods and tools that were deliberated for SPME but has not been conducted any systematic literature review. The main objective of this research is to find out and present the recent research on Software Project Management Education which has been done in recent. These research approaches which different researchers has been conducted and evaluated will lead to the new era of future research. In order to categorize concerned studies in to 5 different classifications: Type of the concerned research, empirical type, contribution type, software project management processes and KA’s, and curricula. A systematic literature review was produced. 82 different research papers have been classified as per defined criteria. The findings of this research are discussed in details and the instructions list obtained from SPM literature for teachers.Item EFFICIENT PREDICTION OF LIVER DISEASE USING SELECTED ATTRIBUTES(UMT, Lahore, 2019) SALAH-U-DIN AYUBILiver plays a vital role in human body that performs several crucial life functions. A number of liver diseases exist and it is a challenging task to diagnose the liver disease at its early stage. In recent years, several data mining techniques have been used for prediction however, there can be further improvements for quick and accurate diagnose of liver disease. Advances in data mining methods for classification and regression open the entrance of recognizing complex patterns from field complex data. In this thesis, a variety of Classifiers have been experimented on Indian liver disease patient’s dataset which is publicly available on Kaggle. Attribute subset selection is performed to identify significant attributes and the resulting dataset is named as Selected Attributes Dataset (SAD). SAD provides more accuracy in less computation time using Random forest classification algorithm. This research work will provide help to predict liver disease with less amount of data, i.e., number of attributes.Item A Study on Critical Success Factors (CSF’s) of Software Development Process, Time and Quality(UMT, Lahore, 2019) SAFDAR IJAZSoftware projects run out of time, exceed from budget, do not meet the customer requirements, and sometime, are canceled before completion. In other words, software development industry is in chaos. With passage of time, software are getting more common but doesn’t seem to be more reliable. Still in 2019 failure rate of software is far more than success rate. According to a survey report about two-third of all information technology (IT) projects get failed due to multiple reasons (Florentine, Feb 2017).To investigate this issue, researcher perform a study by reviewing the literature and then conduct a survey to identify the critical success factors which can lead a software development industry towards the success. The researcher target top IT industries working in Pakistan to find out the critical factors. For this 70 participants, belongs from senior management, executive and senior staff submitted their response. At the end, researcher concluded that there are multiple critical success factors (CSF) that can impact on any software development company. The most underrated success factor identified by this research was Human factor. Synergy, motivation, team work and accountability are the other most important factors in bringing software industry projects to a level of high quality and make the deliverable on time to market within the budgeItem SAFE DIAGNOSIS OF OSTEOPOROSIS USING NON-INVASIVE SIGNALS AND MACHINE LEARNING(UMT, Lahore, 2019) RABIA RIAZOne of the main diseases that affects bone health is Osteoporosis. Osteoporosis causes the bones to become porous and less dense hence decreasing its strength. People having this disease are at higher fracture risk, hence are more prone to break their bones and get fractures. This disease can also cause pain at the joints and certain bones. These patients are likely to have fractures on hip, spine and wrist. It can appear over age or due to the lack of healthy lifestyle and nutrition. Conventionally, osteoporosis is being diagnosed by x-rays, ultrasounds etc. These methods are either radioactive, not readily available or very expensive to be carried. For a small bone checkup in younger age one might has to go through these ways one time or more. Most under developed countries don’t even have this facility in all hospitals. So there is a need to find a way to study bone health in a way that is non-ionizing, less expensive and readily available everywhere. Osteoporosis being a condition that is spread widely all over the world especially in women, needs a way of diagnosis that can be widely and easily used. Diagnosing osteoporosis non-radioactively is a big challenge. Work has been done to diagnose osteoporosis non-invasively by researchers. This thesis intends to find a non-invasive way to study the bone health and diagnose or predict osteoporosis inexpensively and safely. Two different ways are performed to try to achieve the desired non-invasive way of diagnosis. And then their comparison is made to study the effectiveness of the methods. First architecture is traditional way that is already researched on and uses the natural frequency as a parameter of diagnosis and second architecture uses machine leaning. SVM is used as a classifier in architecture two and experiments are performed in two ways. Firstly by performing SVM using grid search and second by performing SVM using cross validation.Item REAL ESTATE MANAGEMENT SYSTEM IN PAKISTAN USING BLOCKCHAIN(UMT, Lahore, 2019) HIFZA SHAUKATHousing schemes are the back bone of Pakistan real estate sector yet they are facing massive problems like misuse as well as pilfering and larceny. The current system of housing scheme within Pakistan is not trustworthy, secure, adequate, and reliable for seller and buyer transactions. To resolve the issue, this research paper purpose a blockchain based transparent, secure, efficient and auditable platform to avoid misuse, illegal transaction of money and give a sense of trust to sellers and buyers. The system and business proposed model is to introduce latest technology of blockchain system as a platform for housing scheme sector in Pakistan which have been provided a secure process of transactions for both seller and buyer. Blockchain is digital peer to peer technology which can provide secure platform for transactions without the requirement of central authority or any third party using interconnected computers, distributed ledgers, consensus mechanism and smart contracts. Blockchain has numerous applications and it can be applied in real estate housing scheme resulting in reduced transaction costs, process and more transparency and accountability in the system. Blockchain system is based on the complete change in business model as the crypto currency will be replaced with fiscal currency, all the buying and selling will be performed in Crypto Currency (CC). The evaluation of performance has been showed the results of behavior of chain increasing on each node and delaying of replicating data on chain in proper time.Item A GENERIC SOFTWARE ESTIMATION MODEL FOR MOBILE BASED BOARD GAMES(UMT, Lahore, 2019) AYESHA FAROOQMobile gaming is an extensive industry which takes over other console equipment. Mobile gaming helps to reduce the size and cost for development and make it more accessible by providing a platform of real time experience to user within the environment. The precise estimation of any software improvement expenses is a basic issue to settle on great management choices and precisely decide how much work and time a task requires for a framework expert, designers and a project manager. Software size and cost estimation are most important phases of development of projects. Software size estimation plays key role for better planning of games development. The main focus of this research article is related to the appraisal of free basis board established mobile games. The major contribution of this research is to evaluate the estimation model for the sizing of mobile games. This model is providing guiding way to developer to check the dimension of software game application to accomplish its work properly. Data were collected and downloaded from different hosting websites. In this research, Single Linear Regression (SLR) and Forward stepwise Multiple Linear Regression (MLR) are used to find out the size estimation of mobile games. These strategies are going to enable a developer to evaluate the game size.Item A POLICY RECOMMENDATION TO RESOLVE E-HEALTH CARE ISSUES THROUGH NON FUNCTIONAL REQUIREMEN(UMT, Lahore, 2019) AREEBA AKBARElectronic Health is basically focused on computer-based techniques or applications for the interaction of patients and doctors through a single electronic environment. The E-Health care system’s development is facing major challenges such as time and cost, language barrier, cultural and norms differences, a difference of states, communication gap, coordination problems, lack of medical project guidance, lack of experienced software testers for health systems, user’s dissatisfaction, poor top project management and so on. These challenges have a major impact on a system’s quality in E-Health and decrease the acceptability among its users. In recent years, some strategies had been applied to reduce quality-related issues that exist in E-Health, however, there is still a need to develop effective and efficient methods, techniques and best practices that can lead to enhance the quality of health software development. The main objective of this study is to highlight existing issues of E-Health and to provide the policy recommendations to mitigate them. Therefore, a quantitative analysis has been performed on collected data to configure the issues of E-Health care system development that directly or indirectly affects the quality of system and services. A detailed study of literature is conducted to test the current proposed framework of E-Health system for security and privacy and to compete for these challenges. It also provides detailed statistical analysis on data by analyzing and applying the demographic profile, frequencies, descriptive studies, reliability test (achieving significance (0.796 > 0.70)), nonparametric test (rejected the null hypothesis with less than 0.5 significance) and one sample T-test (with 2-tailed significance difference 0.00) through SPSS. Hence, this work contributes to making strong policy recommendations to E-Health care system development issues.Item A Lean Approach to Design Use Case for Implementing Block Chain Technology in Pharmaceutical Supply Chain Management Using Hyperledger Fabric Framework(UMT, Lahore, 2019) AHMAD AKBARBlock chain technology removes the central authority by enabling the distributed and decentralized environment. It uses cryptographic principles to make transactions more secure and trustworthy. It is very trendy in different domains to resolves the trust issues and removes the middlemen especially in domain of supply chain management and logistics. In SCM subdomain, like pharmaceutical supply chain management there is huge counterfeit drug problem which is due to lack of coordination, transparency and traceability, which means there is no proper and secure flow of information in pharmaceutical supply chain due to complex processes involved in it. These issues are resolved by the block chain technology, because of its capability to ensure the immutability of digital information. In literature review, there is limited number of unstructured use cases present which shows that there are no proper implementations were seen in pharmaceutical supply chain by using block chain technology. So, it is necessary to develop an approach to design a use case that helps for piloting the block chain solution. This research is based on defining approach to develop a use case for pharmaceutical supply chain by using Hyper ledger Fabric framework. This use case is based on GUEST methodology which provides a standard way to develop use case by integrating with business process model and business strategy. Moreover, technological solution is also designed which includes proposed development environment, network topology, smart contract model, Hyper ledger Fabric based Network model and block chain file system for proposed use case. The piloting of a proposed system and performance evaluation is also done to understand the implementation of use case on small scale in order of processing time, storage required and block size.