International Journal of Interpreting Enigma Engineers (IJIEE)

Lung Cancer Prediction Using Feed Forward Neural Network For Chest Scan Images

2025-12-31T04:04:41+00:00

Abstract—Lung cancer is the leading cause of cancer related death in the world, and increased patient outcomes would lead to improvements in its diagnosis and detection at an early stage. Machine learning (ML) has emerged as a useful instrument in lung cancer diagnosis because it can offer superior diagnosis and aid in treatment decisions. Large clinical data sets, including patient demographics, medical histories, and genetic markers, can be analyzed by ML algorithms to find patterns and associations that can be used to estimate the risk of lung cancer. Numerous machine learning (ML) algorithms, such as logistic regression, support vector machines (SVMs), random forests, and deep learning models, have been used to predict lung cancer. Every method has advantages and disadvantages, and the best one to choose will rely on the particulars of the dataset as well as the intended result. Research has exhibited the effectiveness of machine learning (ML) in the prognosis of lung cancer, with a notable degree of accuracy in recognizing benign and malignant nodules on chest CT scans. This study utilizes deep learning techniques the Xception convolutional neural network model, to precisely classify different forms of lung cancer. To enhance the datasets diversity and boost our models training efficacy we apply data augmentation methods despite the encouraging outcomes, there are difficulties with using ML to predict lung cancer.However, it also comes with several challenges and shortcomings such as the need to select well-chosen and quality data, data bias issues, and the multifaceted decision-making that elaborate machine learning models entail.

Celeste: Deep Learning Based Virtual Try on

2025-12-31T04:31:18+00:00

Nowadays there is a high rate of returns of purchased products in digital shopping due to inability to experience the outfit and as well as lack of efficiency in regular product searching methods. We proposed an AI-powered online platform for buying clothes which combines deep learning and image processing. This platform consists of a try-on module which overlays clothing on user photographs virtually while satisfying the visual accuracy in terms of lighting, shape and texture. The cosine similarity is used for searching a product, which also enables image and text-based search through a VGG based model. The React Native and Expo is used for designing the application for constant cross-platform interaction, whereas MongoDB and Prisma is used to provide high scale data handling. This application enhances user experience with personalized recommendations through virtual try on , reducing the search overhead, and integrating digital convenience and in-store customer engagement.

Customized Path Proposal Utilizing Neural Networks-accepted A* Search Method

2025-12-31T05:35:36+00:00

In this paper, we study a critical problem in spatially based applications: personalised route recommendation, or PRR. Given a road network and users' route queries, the PRR task is to generate user-specific route suggestions. An old-fashioned approach is to tweak search algorithms to yield pathfinding-like results. These methods often focus on narrowing the search space by applying suitable heuristic strategies. Because heuristic strategies for these search algorithms are often built, they are not appropriate for application in scenarios involving complex tasks. Moreover, it is difficult to integrate useful background data into the search procedure. To produce a more principled solution, we propose to apply neural networks to improve search techniques for resolving the PRR issue, which is based on the well-known A* algorithm.

Proquest: System for Career Advancements of The Faculties of Higher Education

2025-12-31T06:15:35+00:00

In contemporary higher education institutions, faculty self-appraisal plays a pivotal role in ensuring quality education and fostering professional development. Decision-making is often slowed down and clarity is diminished by the paperwork-intensive processes used in manual faculty evaluations. The "Automated System for Career Advancements of the Faculties of Higher Education," a web-based platform intended to modernize and expedite the faculty self-appraisal process, is presented in this paper. Faculty members can log and monitor their professional activities, such as research publications, event participation, seminars, and lectures, in real time with the system's user-friendly interface. While an administrative panel provides university administrators with consolidated access to appraisal data for well-informed decision-making regarding faculty development and resource allocation, secure registration and login features guarantee data confidentiality. The suggested solution supports more general objectives of sustainable and paperless administrative practices while improving the effectiveness and transparency of the appraisal process through the use of technology. In the end, it helps to advance high-quality education by fostering a culture of continuous improvement that allows institutions to identify and reward excellence in faculty performance.

Energy Yield Analysis and Design of Solar PV Plants Using Pvsyst

2025-12-31T06:32:14+00:00

This project's goal is to use PVsyst software to estimate solar panels for a chosen location. This project offers a feasibility analysis that uses a PV system module to meet the electricity demands of a specific Nampally, Hyderabad location. A popular renewable energy technology, photovoltaic (PV) solar systems use semiconductor materials to transform sunlight into electricity. By assessing the necessary loads and choosing or determining the appropriate parameters, the modeling is completed. The main goal is to assess how well PV module configurations function in order to increase the output and efficiency of PV power plants operating in various locations. A number of factors are examined and discussed, including load consumption, solar irradiation, atmospheric conditions, and geographic region. PVsyst is a software tool for simulating solar PV systems that is simple, quick, accurate, reliable, and well-founded. Batteries with a broad capacity range of 2.5AH to 20,000AH are chosen with an emphasis on innovation and geographical reach. Additionally, a modular inverter that may be scaled to effectively meet energy demand is desirable. Data source (typical), model (Mono 370wp twin 120 cells), technology (SI-Mono), and load with (370wp 29v).