The K. J. Somaiya College of Engineering (KJSCE) was established in 1983 at Somaiya Vidyavihar, and affiliated with the Somaiya Vidyavihar University. It offers 4 year bachelor's degree courses in the departments of Electronics Engineering, Electronics and Telecommunications Engineering, Computer Engineering, Information Technology and Mechanical Engineering. KJSCE is an autonomous college affiliated to the Somaiya Vidyavihar University spread across approx. 65 acres of posh land and stands among the top 3 engineering colleges in Mumbai. It is also one of the only 7 autonomous engineering colleges in Mumbai the others being VJTI, ICT, SP, KJSIEIT, DJ and Thakur..
Aluminium alloy composites are promising for automotive applications due to their significant properties, including lightweight, enhanced strength, and improved corrosion resistance. However, it found that consequences such as porosity, uneven particle distribution, and agglomerated structure reduce the mechanical behaviour of composites. This research addresses a research gap and enriches the overall properties of the aluminium alloy (AA2024) composite, integrated with 10 wt
The present research investigates the mechanical, fatigue, creep, and dynamic mechanical properties of Aloe vera-polyester-nanocellulose composites to evaluate their structural performance and reinforcement effects. The incorporation of Aloe vera fibers significantly enhanced the mechanical properties of the composite, with further improvements observed upon the addition of nanocellulose. Specimen PAN2, containing 3 vol
This research introduces a cost-effective and portable urine test strip reader tailored for point-of-care diagnostics. By leveraging computer vision and image analysis, the device quantifies colorimetric changes on urine reagent strips, mitigating the limitations of visual interpretation in traditional dipstick methods. The system is built using a Raspberry Pi 4B and a high-definition webcam, measuring the concentrations of 10 crucial urinalysis parameters. It captures images of the test strips at specific intervals, extracting RGB color values from the test patch regions. Calibration curves correlating RGB values with analyte concentrations over time were developed, achieving an accuracy of 88.7
The modern Digital education system is progressively influenced by the usage of Large Language Models (LLMs) as it supports intelligent tutoring, automated assessment, and adaptive learning experiences. Recent LLM models, such as GPT-4, Gemini, and Claude have strong capabilities in reasoning and language understanding; but their usage in education and learning is found to be risky as these models hallucinate information and breaches the academic integrity and safety. A comparative analysis of prominent LLMs is presented in this study. While recent studies address hallucination detection, retrieval-based grounding, or safety evaluation independently, no prior work integrates all three dimensions into a unified evaluation pipeline. To address this gap, this study proposes a Unified Verification Flow based on the LLM-as-Judge paradigm. The framework includes checking for groundedness, finding hallucinations, and aligning safety to systematically test frontier models against specialized academic standards. It shows that, with the usage of LLMs access to learning resources and educational support has tremendously improved. The mechanism to be carried out in the framework is a mandatory check to ensure that something is true. The paper states the necessity of a verify-beforedelivery pipeline that would be used to guarantee the safe and accountable implementation of the technologies of the autopromised LLM within the learning context.
Magnesium-based composites can benefit from a high strength-to-weight ratio. However, agglomeration caused by poor wettability and porosity due to air entrapment and oxidation influences the composite’s behaviour. The addition of 1
