Lahore Garrison University Research Journal of Computer Science and Information Technology
Not a member yet
227 research outputs found
Sort by
Algorithmic as well as Space and Time comparison of various Deep Learning Algorithms
Full text linkDeep learning is an artificial intelligence subfield within machine learning. Now- a-days, deep learning has been used in various applications like computer vision, natural language processing, speech recognition, social network filtering, neural machine translation, etc. Deep learning, Convolutional Neural Network (CNN) is a set of deep neural networks mainly designed for image analysis. Deep learning strong ability is mainly due to multiple feature extraction. In this pa- per, we will discuss and compare AlexNet,VGGNet-16,Residual Network(ResNet-50,101,152)
Secure Cloud Based IoT Data Storage
Full text linkWith the development of emerging technologies, the rate of security and performance of data storage problems are increasing daily and becomes basic necessity of every organizations to store and manage sensitive data in both cloud and IoT environments. Cloud computing is an advanced innovation in which individual can drive work by using several applications. It offers privilege to store data on cloud server while ensuring secrecy of cloud that become a significant problem now a days. Therefore, In the light of previous research study we aims to find various issues and challenges in existing cloud system and provide a better method to store the data in cloud. In our research, we proposed method to improve the secrecy of data storage by using AES encryption with PQC (NTRU) and SHA-512 with modern homomorphic (FHE) cryptographic algorithms. Furthermore, by employing this hybrid technique the performance of cloud storage can be enhance by ensuring the confidentiality and integrity. Furthermore, our proposed method has high performance, throughput and resistance against various attacks to store the large amount of data in cloud
Deep learning to predict Pulmonary Tuberculosis from Lung Posterior Chest Radiographs
Full text linkTuberculosis is one of the most dangerous health conditions on the globe. As it affects the human body, tuberculosis is an infectious illness. According to the World Health Organization, roughly 1.7 million individuals get TB throughout the course of their lifetimes. Pakistan ranks fifth among high-burden nations and is responsible for 61% of the TB burden within the WHO Eastern Mediterranean Region. Various methods and procedures exist for the early identification of TB. However, all methods and techniques have their limits. The bulk of currently known approaches for detecting TB rely on model-based segmentation of the lung. The primary purpose of the proposed study is to identify pulmonary TB utilising chest X-ray (Poster Anterior) lung pictures processed using image processing and machine learning methods. The recommended study introduces a unique model segmentation strategy for TB identification. For classification, CNN, Google Net, and other systems based on deep learning are used. On merged datasets, the best accuracy attained by the suggested method utilising Google Net was 89.58 percent. The recommended study will aid in the detection and accurate diagnosis of TB. 
AN AGRICULTURAL INTERNET OF THINGS (A-IOT) BASED INTELLIGENT SYSTEM FOR DISEASE PREDICTION USING TRANSFER LEARNING, A CASE STUDY
Full text linkThe agricultural Internet of Things (IoT) has altered agricultural output in unprecedented ways. In addition to raising agricultural productivity, it may also significantly raise product quality, lower labor costs, boost farmer income, and achieve agricultural modernization and intelligence. To predict diseases in the agricultural sector, this article proposed an IoT-based smart system for agriculture. The current state of agricultural IoT is first illustrated, along with a summary of its system architecture. In order to predict diseases in the agricultural domain and advance agricultural IoT, an intelligent system based on agriculture is being proposed in this study. The MATLAB 2020a tool is used for simulation and results. In the proposed industrial IoT based intelligent system, a transfer learning model is applied for the training and validation of rice leaf disease prediction in agriculture industry 4.0. The result of the proposed industrial IoT based intelligent system achieved 96.95% accuracy, which is better than the state-of-the-art published methods
NAVIGATING EMOTIONS ACROSS BORDERS: DEEP LEARNING-DRIVEN LOCATION-INFORMED SENTIMENT ANALYSIS OF TWITTER
Full text linkEmotion assessment, a pivotal domain in the realm of natural language processing, holds significant importance in comprehending the feelings expressed through textual content. It has progressed to include the intricate interplay of emotions found in textual information. This paper introduces a pioneering approach to sentiment analysis by amalgamating deep learning techniques and geographic context within the realm of Twitter data. Leveraging an expanded sentiment class set that includes positive, negative, neutral, mixed, ambiguous, happy, sad, angry, fearful, and surprised. Our framework aptly captures diverse emotional expressions. Incorporating location-based sentiment analysis unveils cross-border sentiment dynamics, enriching our understanding of how emotions resonate within various geographical regions.
We present a meticulously designed deep learning model that seamlessly integrates textual content and location information. Through the utilization of text vectorization, embedding layers, and advanced classification techniques, our model achieves exceptional accuracy, F1-score, precision, and recall values. The temporal analysis of tweet timestamps uncovers temporal engagement trends, while the examination of tweet lengths underscores the dynamic range of expression within the Twitter character limitation.
Furthermore, our investigation into the locations reveals Twitter's global presence, with the United States, United Kingdom, and Ukraine emerging as key hubs of activity. This geographical insight augments our comprehension of the platform's diverse user interactions.
This paper not only offers insights into sentiment analysis but also paves the way for future research exploring sentiment dynamics, language variations, and real-time interactions within the Twitter landscape
Machine Vision Approach for Identification of Four Variant Pakistani Rice Using Multi-Features Dataset
Full text linkCrops are the most important and beneficial food source in Pakistan. The demand for food has been an increase in Pakistan due to population growth. Pakistan produced 7,410 million tons of rice according to the financial year survey 2020 (FYS-2020). Pakistani rice has been cultivated in 3,304 hectares of the agricultural land zone, and it is also export around the world. Rice is also increased by 0.6% Gross Domestic Product (GDP) of Pakistan (FYS-2020). The old and manual process of rice classification is more expensive and time-consuming. In this study, we describe a machine vision approach for rice identification. We use four different varieties of rice for the experimental process such as Pakei_Kaynat, Kaynat_Kauchei, and Kauchei_Super_Banaspati and Tootaa_Kauchei (P1, P2, P3, and P4). The 100 images dataset have been used for practical work and total calculated of 400 (4 x 100) image of rice. The different process has been deploying on available datasets such as introduction, preprocessing methodology, and result discussion. A quality enhancement technique has been implementing for clarifying between rice color and shape sampling, and it is also converted color image in gray scale level. Every image has been employing six different non-overlapping regions of interest (ROI’s) and calculated a total of 2400 (6 x 400) ROI’s. Binary (B), Histogram (H) and Texture (T) features have been implemented and extract 43 features on each ROI’s and total calculated 103,200 (2400 x 43) machine learning (ML) features. Best First Search (BFS) Algorithm was used for feature optimization. Different ML classifiers are implementing for experimental process namely; Function Multi-Layer-Perception, Function SMO, Random Tree, J48 Tree, Meta Classifier via Regression and Meta Bagging. The Function Multi-Layer-Perception overall accuracy (OA) has describe better accuracy result is 99.8333%
Requirement Elicitation using Natural Language Processing
Full text link—This paper is the outcome of the research conductedto investigate the affective requirement engineering techniquesproposed and used for developing software projects. We haveassessed traditional methods and proposed an approach thatcovers various aspects for generating a successful project. AnNLP-based model is designed that takes input from the user andgives the output in the form of a text document after processingit. We have set a 62% similarity index to achieve the maximumrequirements of the required system. These requirements, inreturn, help the developers to develop the product with morefunctionality and productivity
A Comparative Study of Parallel and Distributed Big Data programming models: Methodologies, Challenges and Future Directions
Full text linkAccording to a survey conducted in 2021, users share about 4 petabytes of data on Facebook daily. The exponential increase in data (called big data) plays a vital role in machine learning, internet of things (IoT), and business intelligence applications. Due to the rapid increase in big data, research in big data programming models gained much interest in the past decade. Today, many programming paradigms exist to handle big data, and selecting an appropriate model for a project is critical for its success. This study provides an in-depth analysis of big data programming models such as MapReduce, Directed Acyclic Graph (DAG), Bulk Synchronous Parallel (BSP), and SQL. We conduct a comparative study of distributed and parallel big data programming models and categorize these models into three classes: traditional data processing, graph-based processing, and query-based processing models. Furthermore, we evaluate these programming models based on different parameters like performance, data processing, storage, fault-tolerant, suitable language, and machine learning support. Finally, we highlight the benchmark datasets used for big data programming models and discuss the challenges of models along with future directions for the research community
Smarter Garbage Management: CNN with Transfer Learning and Object Detection: Smarter Garbage Management: CNN with Transfer Learning and Object Detection
No full textGarbage is a waste substance that is abandoned by people, generally owing to a perceived lack of utility. We are confronted with the massive amount of garbage generated by people every day that should be properly recycled, reused and repaired by the garbage management system. The first step after garbage collection is to separate or classify garbage into different categories such as glass, paper, plastic, etc. in order to reuse, recycle, repair and recover it. The existing classifiers can only classify garbage in three or six categories. We have designed and implemented a Garbage Classification and Labeling System (GCLS) using SVM and Convolutional Neural Network(CNN) that segregates garbage in eight classes and also label the objects in the image namely cardboard, leather, glass, metal, plastic, paper, rubber and trash. Using transfer learning we have achieved up to 90.4% accuracy that is higher than the existing classifiers
Identification and Classification for Diagnosis of Malaria Disease using Blood Cell Images
Full text linkMachine Learning is a subfield of artificial intelligence that focuses on developing intelligent algorithms capable of learning from available data without requiring constant programming, enabling them to adapt to different environments based on current scenarios. These algorithms are crucial in making intelligent decisions and conducting thorough analyses to uncover intricate patterns concealed within the data. This study used multiple machine-learning classification algorithms to analyze patients' data based explicitly on input images containing parasite-infected and uninfected Malaria samples. AI techniques were utilised to measure the presence of parasites in the images. The image classification system was designed to accurately identify malaria parasites in blood images by generating image features related to color, texture, and cell and parasite geometry. A classifier based on SVM (Support Vector Machine) provided by Weka was employed to differentiate between parasite-infected and non-infected blood images. Through extensive experimentation, it was determined that SVM strategies exhibited significant relevance, achieving a cross-validation accuracy of 99.4% in the basic diagnosis of malaria fever. This finding holds great potential in assisting clinicians with accurate infection diagnoses