Browsing by Author "Majid Ali"

Now showing 1 - 7 of 7
  • Thumbnail Image
    Item
    Application of regression and correlation in real life
    (UMT Lahore, 2022) Majid Ali
    The regression and the correlation is frequently used statistical methods to inspect the relationship between any two or more than two variables. In our life there are many problems in which the correlation and regression help us to solve the problem. Here in this project we have discussed some of the real-life problems and their solutions using correlation and regression. By using these methods, we can solve many different problems of different fields. It is also very helpful in finding the relationship between the stocks of the markets.
  • Thumbnail Image
    Item
    Complete Travel Solutions in Aviation
    (UMT, Lahore, 2022) Majid Ali; Amarullah; Abdul Ahad Khan; Sameer Ahmad; Sheza Kanwal
    The work reveals: that the provision of door-to-door service will assist the passengers and will bring psychological ease along with catering them physically. This research mainly addresses the issue of giving an integrated approach to air passengers. Service includes transport to the airport and from the airport to the final destination by keeping in view the time constraint and consulting with passengers by providing efficient communication by building an application for the record of everything. The low-cost strategy is one of the key features for travel agencies that could be provided to air passengers in assistance.
  • Thumbnail Image
    Item
    Design and implementation of intelligent uninterruptibel power supply (i-ups)
    (University of Management and Technology, 5/23/2013) Majid Ali; Hafiz Zahid Rasool; Majid Rehan Khan
    UPS stands for Uninterruptible Power Supply. It is an instrument connected between the electric grid and the consumer, comprising of electric hardware and rechargeable batteries. The aim of the instrument is to supply continuous undisturbed and conditioned power to the critical load. The energy for powering the load comes from the utility, or from the battery upon mains outage. Uninterruptible Power Supply (UPS) are widely used to provide emergency power to critical loads in case of utility mains failure, and as such constitutes an essential element in providing back-up power for computer networks, communication links, biomedical equipment, and industrial processes, among others. In market local ups designed for just power production using batteries or such types of power sources. They do work but not for all environment. We implement intelligent uninterrupted power supply(i-UPS) for multipurpose. It consists of microcontroller based circuit. Mainly four feature of our i-UPS. First, it is safe and Improvesbattery life by using temperature sensor. When battery is fully charged and comparator do not trip battery charging then microcontroller will come into work trip the battery charging by temperature sensing because heating hardens the battery and it loses its efficiency and life. Second, we provide two outputs from i-UPS, Load-1 and Load-2, Load-1 controls lights, chargers and low power electronics appliances. On the other hand, Load-2 controls the fans and inductor type loads. Third is Energy saving works in the form of battery status. When battery is at its critical situation it switches offLoad-2 and lets the Load-1 on. So we can get lights at low battery status. When main supply is on, automatically Load-1 and Load-2 are connected to main supply. Fourth, we have provided two main energy sourcesLine-1 and Line-2 from different feeders. Microcontroller automatically checks and selects the active Line. If both lines are active then microcontroller will select Line-1 by default setting. This Intelligent Uninterruptible Power Supply (i-UPS) will enable the users to monitor different status of i-UPS on LCD. One of the advantage applying microcontroller for the Intelligent Uninterruptible Power Supply (i-UPS) is that the system is more reliable and user friendly in functions as compare to the conventional Uninterruptible Power Supply available in the market. At the competition of our project we will be hopefully able to save energy and utilization of energy due to intelligent uninterruptible power supply because of our country facing a large amount of electricity crises so people wants such type of ups like i-UPS
  • Thumbnail Image
    Item
    Design and simulation of a demand side management algorithm in a proposed smart grid environment for pakistan
    (University of Management and Technology, 2015) Majid Ali
    The electrical energy cannot be stored in large scale because we have no resource to store it in bulk.It has to be produced, transmitted, distributed and used instantly. However the nature of the load is variable with respect to time. The power generating plants are designed for maximum demand. It means the gap between peak load and the average load is high which results in greater per unit cost. The peak load increases suddenly in peak hours so we need to install large power generating plants to meet peak demand. It is not possible for developing countries to install such large power generating unitsdue to limited financial resources. The proposed solution of this problem is Demand Side Management (DSM) technique. In Demand Side Management (DSM) techniquethe load is shifted to off peak hourswhich results in financial benefit to residential customers.DSM is an important part of Smart Grid(SG)which is used to shift load to off peak timings as well as reshape load curve.As aresult, importance of SG increases and there isa decrease in per unitenergy cost. However this research thesisrevolves around DSM load shifting techniques for only flexible devices. The flexible devices should be shiftedto off peak period. “The Day Ahead Shifting” techniqueis proposed in this research thesis to minimize the above said problem significantly.Mathematical modelling and simulation in this research thesis are based on residential load during the spring season in Pakistan. This algorithm is developed in Matlab software. The findings of this thesissuggest thatDSMapproachattainssignificantenergy savings and objective curve.
  • Thumbnail Image
    Item
    E-learning system
    (University of Management and Technology, 2015) Zeeshan Iqbal; Muzammil Sattar; Noman Ali Saeed; Majid Ali
    E-Learning is the web base system. In this system content is delivered via Internet and helps the students to learn at home. E-Learning web base system has shown its effectiveness now a day's, in education. Almost everyone who use the computer or a smart phone has completed some type of E-Learning. Perhaps it was called tutorial or computer base training and mobile learning applications. In briefed E-Learning system gives worldwide connectivity to the massive knowledge base that provides incremental learning online. E-Learning system can be used in any organization and institute for the sake of knowledge and awareness. In current days E-Learning is the basic need of the every field such as economics, law enforcement, medical and I.T sector. As we mention before it's a web base system so a person he/she can open the site via internet and view all the features of the web site and get benefits from the web site. In this system there are some modules which holds the basic functionality of the web site, Suppose if user visits the site through the browser, he/she can see services, blogs and events, or he/she can be a registered user and use many other features that are restricted to a guest user. For an Individual student, there are different courses against three different packages which are applicable if he/she wants to subscribe a course. There is a specific teacher against a specific course who can provide education online and take the quizzes. Purpose of this project is to create a virtual environment for teachers and students without a class room.
  • Thumbnail Image
    Item
    Investigating the Effectiveness of Few-Shot Learning Approaches for Image Classification on Small Dataset
    (UMT, Lahore, 2025) Majid Ali
    Herbal ingredients are crucial in ensuring effective treatment, patient safety, and the preservation of cultural identity in traditional medicine. However, due to the high visual similarity among different herbs and the limited availability of labeled data, conventional deep learning approaches often struggle with generalization. This paper presents a Few-Shot Learning (FSL) framework to address the challenge of scarce data in herbal image classification. Three established FSL architectures Prototypical Networks, Siamese Networks, and Relation Networks were evaluated and compared with MobileNetV3 as a lightweight backbone for feature extraction. To further enhance performance, a hybrid architecture combining Prototypical Networks and FEAT was developed, leveraging the strengths of each. Extensive data augmentation techniques, including Random Gamma Correction, Auto Augment, and Random Erasing, were applied to improve generalization. A clean data split was used, training on a subset of classes and testing on unseen classes to simulate real-world classification scenarios. Model training followed a 5-way 5-shot episodic format, with evaluation metrics including accuracy, precision, recall, and F1-score. The results show that, compared to traditional FSL models (achieving ~89–98% accuracy on unseen classes), the proposed hybrid model achieved competitive and in some cases superior accuracy (98%), while maintaining feasibility for practical deployment. This study demonstrates that strategic integration of FSL concepts, lightweight architectures, self-attention, and optimized training methods can significantly improve robustness in low-data conditions. The proposed framework holds promise for mobile and edge-based healthcare applications, enabling real-time herbal recognition tools for clinicians, pharmacists, and field researchers
  • No Thumbnail Available
    Item
    Projection of determiner phrase (dp) in punjabi language
    (UMT Lahore, 2018) Majid Ali
    The goal of this thesis is to analyze the syntactic structure of the Punjabi determiner phrase (DP) within the confines of Chomsky’s theory generative grammar. The paradigm of this research is qualitative in nature and it provides in depth account of understanding about the syntactic world. As a result to provide further empirical support to the description of data, the study also employed data analysis by using Chomsky’s theory generative grammar as a model. In order to find out the results focus was given to the routine work and the structure of the Punjabi language. The data for this research was collected from the Punjabi speakers and Punjabi grammar books. The young Punjabi male and female speakers between the age group of 20 to 40 have been taken as a sample of this research. They have been given Punjabi sentences for elicitation in order to analyze the results. In this study, observation was used as a first tool for data collection Moreover, sentences of 30 participants have been observed syntactically. Confidentiality and moral values are very important for this, ethical considerations were taken into account. Four educated M.Phil. Scholars in linguistics as well as language experts have contributed to collect and analyze the data from the sample. The findings of this study also concerned, on the behalf of analysis it transparently clear that the position of the determiner phrase (DPs) in Punjabi language placed likewise the DP structure of English language instead of one point which has been observed In order to test the position of DP at possessive position. It is observed that when noun comes with the possessive quality in the form of “S” genitive it does not appear like the structure of English sentence. This study is limited to the area of Pattoki region district Kasur in order to gain the natural and actual data for the study of the structure of determiner phrases in Punjabi language. Thus the study successfully, described and analyzed the syntactical structure of determiner phrase (DP) in Punjabi language.