Lahore Garrison University Research Journal of Computer Science and Information Technology
Not a member yet
227 research outputs found
Sort by
The Prospects of Computer-Enabled Voting Systems in Pakistan
Full text linkDemocracy is the power vested in people to choose and elect their representatives. However, theprocess of election and voting is prone to rigging leading to undeserving people leading a nationwhich further causes mistrust and agitation amongst the people. Various methods have beenproposed and implemented towards free and fair elections. In this survey we list and discussdifferent methods proposed and adopted for voting. These include the techniques which wereintroduced in past and can be implied in future, the techniques by which voting system can bemade more secure are, the remote voting, internet/online voting, a RFID tags, a fingerprinttechnology and IOT for updating, two languages the extensible markup language and anotherone is the extensible style language are used to design a unique content and cannot be copied.Other technology like electronic voting machine with battery can be useful in rural areas whereinternet is not available and last one is the blockchain technology by which the voter can casttheir vote in no time, and can have trust on that as this technology is end-to-end encrypted, heredata can be saved in sealing blocks. All this work is to find a better way to make the votingsystem more reliable and trustable for the future. In the end we discuss the efficacy of thesesystems in current infrastructure and requirements of Pakistan
Roman Urdu Sentiment Analysis of Reviews on PSL Anthems: Roman Urdu Sentiment Analysis of Reviews on Pakistan Super League Anthems
No full textDue to the easy access of internet and smart devices, people are becoming habitual to give their feedback on what they hear or watch, online. These reviews are very valuable for all sorts of users. Due to the widespread online activities, the count of these reviews has raised tremendously. This fact makes it humanly impossible to analyse them manually. So it needs time that reviews to be analysed and use patterns to be found and explored through the automated channel. This led to a new field of research known as Sentiment Analysis. This paper is targeting to design a model to perform sentiment analysis of Roman Urdu text using the reviews of Pakistan Super League’s official song. To perform this analysis five different techniques-- Naïve Bayes Kernal, Random Forest, Logistic Regression, K-Nearest Neighbour and Artificial Neural Network, are applied. Naïve Bayes Kernal and Logistic Regression correctly predicted 97.00% reviews
A Predictive Analysis of Retail Sales Forecasting using Machine Learning Techniques
Full text linkIn a retail industry, sales forecasting is an important part related to supply chain management and operations between the retailer and manufacturers. The abundant growth of the digital data has minimized the traditional system and approaches to do a specific task. Sales forecasting is the most challenging task for the inventory management, marketing, customer service and Business financial planning for the retail industry. In this paper we performed predictive analysis of retail sales of Citadel POS dataset, using different machine learning techniques. We implemented different regression (Linear regression, Random Forest Regression, Gradient Boosting Regression) and time series models (ARIMA LSTM), models for sale forecasting, and provided detailed predictive analysis and evaluation. The dataset used in this research work is obtained from Citadel POS (Point Of Sale) from 2013 to 2018 that is a cloud base application and facilitates retail store to carryout transactions, manage inventories, customers, vendors, view reports, manage sales, and tender data locally. The results show that Xgboost outperformed time series and other regression models and achieved best performance with MAE of 0.516 and RMSE of 0.63
Deep Reinforcement Learning for Control of Microgrids: A Review
Full text linkA microgrid is widely accepted as a prominent solution to enhance resilience and performance in distributed power systems. Microgrids are flexible for adding distributed energy resources in the ecosystem of the electrical networks. Control techniques are used to synchronize distributed energy resources (DERs) due to their turbulent nature. DERs including alternating current, direct current and hybrid load with storage systems have been used in microgrids quite frequently due to which controlling the flow of energy in microgrids have been complex task with traditional control approaches. Distributed as well central approach to apply control algorithms is well-known methods to regulate frequency and voltage in microgrids. Recently techniques based of artificial intelligence are being applied for the problems that arise in operation and control of latest generation microgrids and smart grids. Such techniques are categorized in machine learning and deep learning in broader terms. The objective of this research is to survey the latest strategies of control in microgrids using the deep reinforcement learning approach (DRL). Other techniques of artificial intelligence had already been reviewed extensively but the use of DRL has increased in the past couple of years. To bridge the gap for the researchers, this survey paper is being presented with a focus on only Microgrids control DRL techniques for voltage control and frequency regulation with distributed, cooperative and multi agent approaches are presented in this research
Descriptive Analysis of Human Emotions Based on Eye pupils
Full text linkFacial emotional expressions are viewed as the most descriptive way to understand the human’s state of temperament during confronting communication. In this work numerous statistical approaches have been applied on human eye pupil with static images of Chicago face dataset (CFD) to analyze and classify the considered categories for emotions which are Happy, Fear, Anger and Neutral. The aim of this study is to develop the specific architecture for image processing domain after applying different enhancement techniques on human eye pupil for analysis & recognition of the facial expressions. This work is divided into three phases initially in the first phase data preprocessing is performed to prepare according to the requirement of work and also the color images are converted in to negative by applying the pixel intensity controlled mechanism. Second phase define the boundary to compute the feature by using Circular Hough Transform algorithm. Lastly statistical approaches are applied on extracted features to corporate the central point of pupil. This corporation the central point presents the effects of emotions. While comparing peoples of different Age groups it is concluded that pupil constricted on Anger at different levels on different age groups. If further it is discussed about cross cultural and gender wise comparison then Happy Emotion effects most and resulted towards dilated pupil same like that Anger emotion effects most on constricting the pupil size
Comparative Analysis of Machine Learning Techniques for Predicting Air Pollution
Full text linkThe modern and motorized way of life has cultured air pollution. Air pollution has become the biggest rival of robust living. This situation is becoming more lethal in developing countries and so in Pakistan. Hence, this inquiry was carried out to propose an architecture design that could make real-time prediction of air pollution with another purpose of scanning the frequently adopted algorithm in past investigations. In addition, it was also intended to narrate the toxic effects of air pollution on human health. So, this research was carried out on a large dataset of Seoul as an adequate dataset of Pakistan was not attainable. The dataset consisted of three years (2017-2019) including 647,512 instances and 11 attributes. The four distinctive algorithms termed Random Forest, Linear Regression, Decision Tree and XGBoosting were employed. It was inferred that XGB is more promising and feasible in predicting concentration level of NO2, O3, SO2, PM10, PM2.5 and CO with the lowest RMSE and MAE values of 0.0111, 0.0262, 0.0168, 49.64, 41.68 and 0.1856 and 0.0067, 0.0096, 0.0017, 12.28, 7.63 and 0.0982 respectively. Furthermore, it was found out as well that the Random Forest was preferred mostly in the previous studies related to air pollution prophecy while many probes supported that air pollution is very detrimental to human health especially long-lasting exposure causes lung cancer, respiratory and cardiovascular diseases
Impulse Noise Removal Using Soft-computing
Full text linkImage restoration has become a powerful domain now a days. In numerous real life applications Image restoration is important field because where image quality matters it existed like astronomical imaging, defense application, medical imaging and security systems. In real life applications normally image quality disturbed due to image acquisition problems like satellite system images cannot get statically as source and object both moving so noise occurring. Image restoration process involves to deal with that corrupted image. Degradation model used to train filtering techniques for both detection and removal of noise phase. This degeneration is usually the result of excess scar or noise. Standard impulse noise injection techniques are used for standard images. Early noise removal techniques perform better for simple kind of noise but have some deficiencies somewhere in sense of detection or removal process, so our focus is on soft computing techniques non classic algorithmic approach and using (ANN) artificial neural networks. These Fuzzy rules-based techniques performs better than traditional filtering techniques in sense of edge preservation
The Estimation of outliers in cognitive networks spectrum sensing: Estimation of outliers in cognitive networks spectrum sensing
No full textThe choice of this topic was influenced from the concept that statistical analysis of different attributes representing certain endpoints of behavior during radio communication in cognitive networks was necessary to study the outliers occurring in those parameters. The importance of cognitive radio is explained in detail in the literature review section of this paper. The purpose of this report is to do an overview of emerging patterns in cognitive radio networks and seek an understanding of data by learning what kind of attributes that display outliers during estimation. During the course of this research, it has come to light that study of outliers require preprocessing of data during which certain anomalies of data are studied and then removed thus optimizing the dataset. In the process, two major attributes SNR and Lambda have emerged and statistically shown a pattern that helped with the estimation of outliers.
Key words: SNR, Lambda, Outliers, PU, SU, CRs
MM-Wave HetNet in 5G and beyond Cellular Networks Reinforcement Learning Method to improve QoS and Exploiting Path Loss Model
Full text linkThis paper presents High density heterogeneous networks (HetNet) which are the most promising technology for the fifth generation (5G) cellular network. Since 5G will be available for a long time, previous generation networking systems will need customization and updates. We examine the merits and drawbacks of legacy and Q-Learning (QL)-based adaptive resource allocation systems. Furthermore, various comparisons between methods and schemes are made for the purpose of evaluating the solutions for future generation. Microwave macro cells are used to enable extra high capacity such as Long-Term Evolution (LTE), eNodeB (eNB), and Multimedia Communications Wireless technology (MC), in which they are most likely to be deployed. This paper also presents four scenarios for 5G mm-Wave implementation, including proposed system architectures. The WL algorithm allocates optimal power to the small cell base station (SBS) to satisfy the minimum necessary capacity of macro cell user equipment (MUEs) and small cell user equipment (SCUEs) in order to provide quality of service (QoS) (SUEs). The challenges with dense HetNet and the massive backhaul traffic they generate are discussed in this study. Finally, a core HetNet design based on clusters is aimed at reducing backhaul traffic. According to our findings, MM-wave HetNet and MEC can be useful in a wide range of applications, including ultra-high data rate and low latency communications in 5G and beyond. We also used the channel model simulator to examine the directional power delay profile with received signal power, path loss, and path loss exponent (PLE) for both LOS and NLOS using uniform linear array (ULA) 2X2 and 64x16 antenna configurations at 38 GHz and 73 GHz mmWave bands for both LOS and NLOS (NYUSIM). The simulation results show the performance of several path loss models in the mmWave and sub-6 GHz bands. The path loss in the close-in (CI) model at mmWave bands is higher than that of open space and two ray path loss models because it considers all shadowing and reflection effects between transmitter and receiver. We also compared the suggested method to existing models like Amiri, Su, Alsobhi, Iqbal, and greedy (non adaptive), and found that it not only enhanced MUE and SUE minimum capacities and reduced BT complexity, but it also established a new minimum QoS threshold. We also talked about 6G researches in the future. When compared to utilizing the dual slope route loss model alone in a hybrid heterogeneous network, our simulation findings show that decoupling is more visible when employing the dual slope path loss model, which enhances system performance in terms of coverage and data rate
A Formal Model for Smart Living Room
Full text linkwe are living in an era full of technology and the most powerful feature behind this technology is the communication between two or more things. We achieved globalization with the power of digital computers and their ability to communicate. The next shape of computers for interactive remote processing is internet of things or wireless sensors network and for data storage it is cloud. These tiny computers with heterogeneous characteristics are very helpful in making environment smart and interactive in different ways. In this paper, we are proposing an Ambient Intelligence architecture for safety and energy efficiency using sensors, further we are formalizing the architecture for its accuracy and reliability. The three major sensors are smoke sensor for safety, glass break detector sensor for security, motion sensor for energy efficiency. In addition, the working of all sensors is also formalized for its correctness