International Journal of Engineering and Management Research
Not a member yet
1311 research outputs found
Sort by
Enhancing Spammer Fake Profile Detection on Social Media Platforms using Artificial Neural Networks
Get PDFThe proliferation of social media platforms has led to an increase in spammer fake profiles, posing significant security, privacy, and trustworthiness concerns. Traditional manual monitoring and content filtering techniques are insufficient to combat this growing issue, necessitating the development of more efficient and accurate detection methods. Machine learning techniques have been increasingly employed for this purpose, demonstrating promising results in identifying spammers and fake profiles. This paper presents a novel approach for spammer fake profile detection using Artificial Neural Networks (ANNs) to enhance the accuracy of the detection process. Our proposed ANN-based method addresses the challenges associated with spammer fake profile detection, such as the dynamic nature of spammers, data heterogeneity, scalability, and imbalanced datasets. We evaluate the performance of our method on real-world datasets and compare it with existing machine learning techniques, demonstrating its effectiveness and superiority in detecting spammers and fake profiles with higher accuracy. This research contributes to ongoing efforts to secure social media platforms, ensuring the trustworthiness of online content and providing a safer user experience
Prediction of Loan Approval in Banks using Machine Learning Approach
Get PDFDue to significant technology advancements, people\u27s needs have expanded. As a result, there have been more requests for loan approval in the banking sector. A few qualities, taken for consideration, when choosing a candidate for loan approval in order to, determine loan\u27s status. Banks face a major challenge; when it, comes to assessing loan applications and lowering the risks associated with potential borrower defaults. Since they must thoroughly evaluate each borrower\u27s eligibility for a loan, banks find this process to be particularly challenging. This research proposes combining machine learning (ML) models and ensemble learning approaches to find the probability of accepting individual loan requests. This tactic can increase the accuracy with which qualified candidates are selected from a pool of applicants. As a result, this method can be used to address the problems with loan approval processes outlined above. Both the loan applicants and the bank employees profit from the strategy\u27s dramatic reduction in sanctioning time. Because of the banking industry\u27s expansion, more people were applying to loans at banks. In order to predict the accuracy of loan approval status for applied person, we used four different algorithms namely Random Forest, Naive Bayes, Decision Tree, and KNN. By using these, we obtained better accuracy of 83.73% with Naïve Bayes algorithm as best one
Bird Flu Tracking System using Naive Bayes Classifier
Get PDFComplexity in the food-related Supply chains also introduces economic inefficiency at every point of transaction, be it a poultry farmer or a distributor. Therefore, a system is required that tracks every point of Poultry transaction with an intuitive rather than minimal technology medium that can trace bird disease and maximize efficiency in operations at every level, from Poultry farm to Chicken Butcher/seller. A solution for farm-to-butcher/seller chicken tracker that rapidly traces bird flu while also providing a platform to maximize operational efficiency. To make the chain effective the proposed system needs enrolment at every level (Farmers, Distributors, Sellers) where poultry transactions can be recorded, and hence, our solution maximizes platform enrolment by cutting down the distributors’ labor costs by providing access to accounting and management tools and providing a one-stop marketplace to both distributors and farmers. With the ease of use of the RFID tagged system for distributors and sellers to tap and add batches to their inventory for easy supply chain management. The same data will also help in flu detection
An Analysis of the Effectiveness of Different Types of Lids in Managing Urban Storm Water
Get PDFCity increase is one such phenomenon that drives deep changes in land use patterns. From a hydrological perspective, urbanization will increase the impervious surfaces, which end result lower of infiltration and boom of runoff. This consequences is a boom in runoff extent and glide that could reason flooding, watercourse and habitat destruction. In recent state of affairs several Indian cities have witnessed remarkable incidences of flooding due to numerous reasons. Defective urban planning and failure of drainage gadget are taken into consideration as principal motives in the back of flooding and surface inundation in urban regions in growing countries. Most important question stand up that how runoff from new tendencies must be controlled? For that the use of SWMM (Storm Water Management Model) for specific Low Impact Development (LID) as gear for lowering runoff as well as discharge in the storm drainage
CodePlex: Software Complexity Measuring Tool based on ECB Measure
Get PDFThe surge in popularity of object-oriented programming as a predominant paradigm in software development has spurred numerous studies to introduce metrics for assessing the complexity of object-oriented programs. These metrics typically fall into two primary categories: those focusing on object-oriented aspects and those centered on cognitive aspects, delineating their principal areas of concern when evaluating program complexity. Within the realm of cognitive aspects, the majority of metrics have historically been confined to the consideration of no more than three complexity variables. However, the ECB (Enhanced Cognitive Based) measure stands as a notable exception, capable of encompassing and addressing four or more intricate facets in the assessment of software program intricacy and difficulty. This research paper undertakes the exploration of the incorporation of these multidimensional metrics as refinements to the existing weighted composite complexity CB measure, originally introduced by Chhillar and Bhasin. In doing so, it endeavors to furnish a more comprehensive and holistic framework for the evaluation of program complexity, accommodating both object-oriented and cognitive dimensions. Furthermore, the study assumes the pivotal role of empirically validating the practical effectiveness of the ECB measure, seeking to bridge the chasm between theoretical metrics and their tangible applicability in real-world settings. Such an endeavor holds profound significance for software developers and researchers, proffering invaluable insights that can advance our understanding and management of intricate object-oriented programs
A Deep Dive into Deep Learning
Get PDFDeep learning is a technique for mimicking the human brain by intimating human functionality and attempting to uncover fruitful patterns in data using a neural network. With the increase in data volume, deep learning is becoming more popular. End devices such as smartphones and IoT sensors generate data that must be appropriately analyzed using deep learning models. This paper intends to present the reader with complete understanding of the fundamentals of deep learning elements in order to make the principles more evident in the deep learning field. This study focuses on the three major types of neural networks that serve as the foundation of deep learning models. The three primary types are as follows: i) Artificial Neural Network, ii) Convolution Neural Network and iii) Recurrent Neural Network. Let\u27s take a deep dive into each of these sorts
The Significance of AI Enhanced Customer Feedback for Providing Insights on Customer Retention and Engagement Strategies for Mobile Companies
No full textAn increasing number of people are interested in learning how artificial intelligence may improve the effectiveness of the company\u27s automated service encounters with its customers. This study aims to lay the foundation for a theoretical framework that will explain how businesses and customers may use AI-enabled information processing systems to enhance the outcomes of both requested and unrequested types of online customer engagement. The goal of this essay, which utilises a Stimulus-Organism-Response theory paradigm and leverages the concept of AI systems as organisms, is to distinguish between requested and uninvited online consumer contact behaviours. These acts cause AI creatures to analyse customer data, resulting in responses from both computers and people that modify the circumstances of subsequent online interactions with these groups. The advent of both digital advertising and AI in the last few decades has had far-reaching effects on the global economy. Artificial intelligence (AI) is already embedded in many aspects of our society. It is also anticipated that marketing and public relations would make heavy use of AI. New product suggestions based on predicted user behaviour could be generated by this system. In addition, it simplifies the analysis of massive amounts of data, which would be extremely difficult to handle by hand. It can also be used to forecast consumer opinion about a company and its products in general. This study\u27s overall objective is to outfit a total structure for directing portable media research tasks to support specialists\u27 mission of productive versatile promoting
Incentive Pay Plan for Improving Safety Performance of Seaport Employees
Get PDFSafety is a proactive, continuous, systemic, and comprehensive process which focuses on reducing risk, injuries, and fatalities in work. Seaports have still higher rate of injuries and therefore innovative initiatives to reduce accidents and improving safety conditions are needed at seaports. Chittagong Seaport is one of the busiest seaports in the world. Due to the complexity and forceful nature of work, management of health and safety may not continuously be up-front in CPA. This study focuses on the health and safety aspects of the Chittagong seaport labor system. In this paper, the incentive as an approach to motivate port employees in using safety equipment was assessed. After applying this method in Chittagong Seaport, there is a behavioral change among the employees. It has been found from the result that there is a significant relationship between incentive and safety performance. The result of this study showed that how incentive has a positive impact in enhancing safety performance of port employees. The use of incentives can be helpful for seaport’s labors to be associated with safety programs
Key Management Strategy and Distribution of Public Key for Cloud Security
Get PDFThe security perspective of cloud systems has developed Hypertext transfer protocol secure (HTTPS), Use case of determination of the public key is widely used in Transport Layer Security for the system. Three components, including storage servers, data storage, and blocking stockpiling, were used to categorise online storage. The limitation is that the waiting period falls inside the time window j > j in order to guarantee advance assurance. The constants for the honeycomb techniques must fulfil the condition that m 2ndlog QE in order to ensure the integrity of the q-module arithmetic lattices. Asymmetrical data cryptography and bilateral cryptography are indeed the two types of key decryption techniques used in authenticating methods. The main goal of encryption is to create secure data for communication links
AI Chatbot for College Enquiry
Get PDFEvery campus needs a campus guide to answer different types of questions without any problem. Every newly enrolled student in the college has a long list of queries- “What are college timings”, “Where is the gym?”, “When does the library open?”, etc. Further, students have queries regarding syllabus, academic calendar, semester break and so on and so forth.To tackle these problems we are incorporating Artificial Intelligence(AI) technology which is spreading wide across all the fields, and reduces the time and efforts of humans by providing optimized results. So, we are going to implement a virtual assistant using artificial intelligence techniques and natural language processing that can solve any college related query. This will function as a machine with Artificial based intelligence. Machine learning algorithms will be used to train the bot, and once trained on a large enough dataset, the bot will understand college-related inquiries. The bot also understands the intent and meaning behind the orders. The AI chatbot identifies the input, context, and intent, which then reacts accordingly