Browse
Recent Submissions
Item ICROTONYLK-PSEAAC(UMT, Lahore, 2019) MUHAMMAD SAFI UR REHMANAmong different Post-Translational Modification (PTM) the most vital one is lysine Crotonylation 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 Crotonylation along with their site is to completely apprehend the mechanism behind this biological process. In the previous research models, 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 to compute the accuracy, sensitivity, specificity and MCC using 10-fold cross validation. The results of independent dataset testing were 99% accuracy, 99.4% sensitivity, 89.1% specificity and 0.98 MCC. Our model have given more accuracy than other research models, using ANN.Item Evaluation of EMR software packages based on integrated AHP, Topsis and fuzzy-Topsis(UMT, Lahore, 2019) SYEDA NISAR FATIMAOpen Electronic medical Records (EMR) systems have gained significant importance in medical field. Evaluation and selection of software packages that meet the specifications of an organization are challenging issue of software engineering procedure. Inappropriate selection of OSS package can be costly and have destructive effects on business processes and smooth functioning of the organization. The purpose of this research is to evaluate and select OSS packages using multi-criteria decision making techniques. Comprehensive analysis on multi-alternatives and multi-criteria conditions was carried out to inspect the OSS package more closely to find a best alternative. Various evaluation measures were identified and the system was selected based on the metric outcomes using Analysis Hierarchy Process (AHP), Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) and Fuzzy TOPSIS. The after effects of MCDM techniques demonstrated that Open EMR software is the best available open-source softwareItem Impact of automation on the performance of revenue collection departments(UMT, Lahore, 2019) MUHAMMAD FAISAL SHAFIQIn today’s fast paced business environment, the maximum use of available resources is not an option but rather a requirement. The departments in public sector are also using proactive approach for optimal uses of their resources through automations of business processes. The Excise & Taxation department is a leading revenue collection department in Punjab. The Government always has increasing need of revenue collections to meet their fiscal requirements. To achieve the revenue targets, it is need of the time to modernize the revenue collection process through automation. This study focused on assessing the impact of automation of revenue collection processes in ET&NC. This study also determined that in what ways automation of tax collection helped to broaden the tax base, and suppress the leakages of taxes. For this study the revenue collection data under the levy of Property Tax of Excise, Taxation & Narcotics Control Department, Punjab was used for the period from July, 2013 to June, 2017. Quantitative data was analyzed by descriptive method and content analysis was used for qualitative data. This research established that automation of tax processes in ET&NC department has great impact both in the form of revenue collection as well as in the shape of better service delivery. This study also recommended that after the automation of all levies, the online tax collection system should be implemented in collaboration with banks for the purpose of easily tax payments.Item Prediction of osteoporosis through deep learning with the help of MRI(UMT, Lahore, 2019) Mazhar Javed AwanOsteoporosis is a disease that weakens the bone strength causing fractures and life-threatening complications. It has been estimated to affect more than 200 million people worldwide. Improving diagnostic technology and assessment facilities have made it possible to detect the disease before it causes fractures. Early prediction of fracture risk enables preventive actions and lifestyle changes that can improve the patient’s quality of life and save costs to society. Diagnosis of osteoporosis is based on measuring the bone density using MRI. This technique produces an image representing bone density at the scanned site. However, bone density itself is only a moderate predictor of fracture risk, which creates demand for alternative prediction models. Machine learning, and especially convolutional neural networks, has been the leading image analysis approach in recent years. It has produced good results also in medical image analysis, including some orthopedically problems. This study seeks to discover if convolutional neural networks can predict osteoporotic fractures from spine MRI images. By experimenting with different network architectures, the study aims to gain an understanding of the most promising design directions of such prediction models. In this context, also some practical machine learning challenges such as low training speed and lack of transparency are addressed.Item Government revenue enhancement through EIS integration and intelligent monitoring(UMT, Lahore, 2019) MUHAMMAD SALEEMThis is a proven fact that tax revenue has a vital significance in the progress and growth of any country. Adopting the same ways in Pakistan, the property tax is collected by Excise, Taxation& Narcotics Control Department (ET&NCD), Government of the Punjab, contributing a second top significant share in total revenue collections of the Department. Currently ET&NC Department using MIS & GIS systems for collection of property tax separately, but details about the properties are still entered manually causing some pilferages to Government exchequer and unable to meet the collection according to its potential demand. This research proposes an intelligent integrated GIS and MIS based system that enables the tax managers to monitor and regulate the property tax collection in an enhanced way. The essential scheme of this particular system devises the escalation of Government Revenue. Certainly, specific endeavor are required for the integration of Enterprise Information Systems (EIS) with Geographic Information System (GIS). The GIS can be implemented using Google Street View (GSV), Google Earth and Google 3D Building Modeler and existing Geo Mappings/Parcels. Moreover, the system can be integrated with Real Estate Agents and District Police system for identification and marking of rental properties. The existing MIS is a joint venture and applied on Property Taxable Circles/Areas in Punjab. This will identify the discrepancy/variation in existing MIS for Property Taxation and prompt & record this discrepancy according to its magnitude. Discrepancy or pilferages in Property Taxation can be minimized by this proposed system and play a key role to collect difference of revenue. The data has been extracted using ETL (Extract, Transform and Load) and a web service (API) from existing available database. The data extraction exhaustively cohere with the business logics of GIS based area determination algorithms. This research thesis, basically, focuses on revenue enhancement by integrating existing Property Tax MIS, GIS & allied technologies, District Police System for Temporary Residence Act and Real Estate Agents System. Researchers believes that proposed research methodology will enhance recovery of property tax from 15 to 20 % annually leading to almost Rs. 2600 million.Item DNAPred_Prot(UMT, Lahore, 2019) Syed Ali HassanIn the field of genome annotation identification of DNA binding protein is one the critical challenge. DNA act as a blueprint for the cell in which all necessary information for building and maintaining the trait of an organism. It is DNA which makes the living thing, a living thing. Protein interaction with DNA performs a vital role in regulating DNA functions such as DNA repair, transcription and regulation. Identification of these proteins is an essential task for understanding the regulation of genes. Several methods have been developed to identify the binding sites of DNA and protein depending upon the structures and sequences, but they were costly and time-consuming. Therefore, we propose a methodology named "DNAPred_Prot", which makes use of features gained from position-relative-incidence-matrix (PRIM) that helps in training for efficient and effective prediction of DNA-binding proteins. Using testing techniques like 10-fold cross-validation and jackknife testing an accuracy of 94.95% and 95.11% was yielded respectively. The robustness of the model has been tested by using independent dataset PDB186 and an accuracy of 91.47% achieved by it. From these results, it can be predicted that suggested methodology performs better than other extant methods for identification of DNA-binding Protein.Item A framework to secure health monitoring applications on android platform(UMT, Lahore, 2019) SOBIA MEHRBANSmartphones in combination with medical software have made documentations, record keeping, have access to information and inter-personnel communication which is faster and more effective in the field of mobile health. Healthcare monitoring applications(HMA) are the applications of software and its work at the level of hardware, as HMA are mostly used to sort out chronic diseases, monitor the elder and expectant mothers, and help the patients to schedule the medicine time, and provide effective and efficient service to the patients, doctors and caretakers. Whereas, many operating system are present in smartphones nowadays, however, Android is the most popular operating system and it handled all the sensitive data that are useful for the professional (doctor, physician, and consultant) over unsecured internet communication and hacker’s server. Due to this, treatment involves basic privacy issues regarding different data of patients, doctors and caregivers. As security in health monitoring apps is necessary to keep the patient’s records confidential, however, their data can be exchanged with wireless networks, and this is the main threat in health monitoring applications. Thus, the important goal is to provide proper security for the patient’s data from unauthorized people who actually misuse the health records. The major concern is to protect all the patient’s sensitive data from unauthorized third parties. In this thesis a secure real time framework has been presented which will help to secure the medical data related with Android health monitoring apps. Through this framework the medical data can be stored properly, protected from threats and improved records monitoring abilities. The proposed security framework has been designed consisting of two layers; one is, A Monitoring Layer (ML) which is used to enforce the modules of security-checks, and second is, an Interface Layer (IL) which provide interface between Monitoring Layer and Android Operating System. ML implements different polices at several levels of android platform through IL. The validation of framework is shown through the implementation of actual Android device, and this implementation shows the framework and evaluate its performance practically. Different experiments have been performed to check the efficiency and effectiveness of proposed framework, and the results of experiments have successfully protect the health monitoring applications against different attacks.Item Methods and tools for database education(UMT, Lahore, 2019) Zeeshan ArshadDatabase System (DS) has got a great consideration from analysts and specialists in previous years. Database System Education (DSE) is a core field for those competent who are able of effectively achieving database projects. As a result of increasing demand, universities should provide computer science students with the dense groundwork in subject matter. There are many unique methods and tools that were discussed for DSE but has not been conducted any systematic literature review. This paper emphases to classify the recent research on DSE that is available till present. This indicates the useful methodologies, identifies the needs that are useful for further future research. In order to categorize the selected study into following classification principles: type of research, type of empirical, type of contribution, DS activities and curricula, a systematic mapping study produced. Total 48 articles were nominated and classified according to the above criteria. Finally, the results are debated and a list of advices are obtained from DSE literature for teachers.Item Intelligent control of hybrid renewable energy sources and integration with smart grid(UMT, Lahore, 2019) Ch. Taimour AhmedElectrical energy is the cleanest type of quick usable energy. Its unique form of transportation and generation has provided jobs to many young engineers and common skilled and unskilled people. This work form a basis of Hybrid renewable energy resources including electrical energy resources combined with battery, considered the excellent source of electric power energy. So that we can estimate the need of generation requirements of our country. This research work will definitely help to minimize the burden on conventional power systems and it will aid something new to existing system.Item A survey on the role of IoT technology in agriculture for the implementation of smart farming(UMT, Lahore, 2019) SHAMYLA RIAZInternet of things (IoT) is a promising technology which provides efficient and reliable solutions towards the modernization of agriculture. IoT agricultural solutions have developed in order to maintain and monitor the whole environment automatically with minimum human involvement. This research surveys the state of the art in IoT agricultural technologies and reviews IoT agricultural applications, sensors/devices, communication protocols. In this thesis also proposed an IoT agricultural network architecture, platform and topology to evaluate the current status of IoT in smart farming. In addition, this thesis also analyzes IoT agriculture security threats including threat model, security requirements, attack taxonomies, and proposed a collaborative security model to minimize the security risks. Besides, we have discussed how IoT agricultural technologies and industries trends have leveraged in a smart farming context. Furthermore, we discuss the IoT Agricultural regulations and policies adopted by different countries in order to measure how they facilitate the farmers and agriculturists in terms of acceptable and sustainable development. Finally, we have identified open research issues and challenges in IoT agriculture field.