MIT Academy of Engineering

Recommender systems or recommendation systems are a subset of information filtering system that used to anticipate the 'evaluation' or 'preference' that user would feed to an item. In recent years E-commerce applications are widely using Recommender system. Generally the most popular Ecommerce sites are probably music, news, books, research articles, and products. Recommender systems are also available for business experts, jokes, restaurants, financial services, life insurance and twitter followers. Recommender systems have formulated in parallel with the web. Initially Recommender systems were based on demographic, content-based filtering and collaborative filtering. Currently, these systems are incorporating social information for enhancing a quality of recommendation process. For betterment of recommendation process in the future, Recommender systems will use personal, implicit and local information from the Internet. This paper provides an overview of recommender systems that include collaborative filtering, content-based filtering and hybrid approach of recommender system.

Cancer has identified a diverse condition of several various subtypes. The timely screening and course of treatment of a cancer form is now a requirement in early cancer research because it supports the medical treatment of patients. Many research teams studied the application of ML and Deep Learning methods in the field of biomedicine and bioinformatics in the classification of people with cancer across high- or low-risk categories. These techniques have therefore been used as a model for the development and treatment of cancer. As, it is important that ML instruments are capable of detecting key features from complex datasets. Many of these methods are widely used for the development of predictive models for predicating a cure for cancer, some of the methods are artificial neural networks (ANNs), support vector machine (SVMs) and decision trees (DTs). While we can understand cancer progression with the use of ML methods, an adequate validity level is needed to take these methods into consideration in clinical practice every day. In this study, the ML & DL approaches used in cancer progression modeling are reviewed. The predictions addressed are mostly linked to specific ML, input, and data samples supervision.

Liver disease diagnosis is a major medical challenge in developing nations. Every year around 30 billion people face liver failure issues resulting in their death. The past liver abnormality detection models have faced less accuracy and high theory of constraint metrics. The lesion on the liver hasn't been identified clearly with earlier models, so an advanced, efficient, and effective liver disease detection is essential. To overcome the limitations of existing models, this approach proposes a deep liver abnormality detection with DenseNet convolutional neural network (CNN) based deep learning technique. This work collected liver Computed Tomography (CT) scan images from Kaggle dataset for training in the initial stage. The pre-processing has been performed with region-growing segmentation, and training is performed through DenseNet CNN. The real-time test images are collected from Government General Hospital Vijayawada (10,000 samples), verified on proposed DenseNet CNN to diagnose whether the input has a liver lesion. Finally, the results obtained and derived confusion matrix summarizes the performance of the proposed methodology with following metrics of accuracy at 98.34%, sensitivity at 99.72%, recall at 97.84%, throughput at 98.43% and detection rate at 93.41%. The comparison results reveals that the proposed technique attains more accuracy and outperforms the other pioneer methodologies.

The most well-known and widely used non-traditional manufacturing method is electrical discharge machining (EDM). It is well-known for its ability to cut rigid materials and high-temperature alloys that are difficult to machine with traditional methods. The significant challenges encountered in EDM are high tool wear rate, low material removal rate, and high surface roughness caused by the continuous electric spark generated between the tool and the workpiece. Researchers have reported using a variety of approaches to overcome this challenge, such as combining the die-sinking EDM process with cryogenic treatment, cryogenic cooling, powder-mixed processing, ultrasonic assistance, and other methods. This paper examines the results of these association techniques on various performance measures, such as material removal rate (MRR), tool wear rate (TWR), surface roughness, surface integrity, and recast layer formed during machining, and identifies potential gap areas and proposes a solution. The manuscript is useful for improving performance and introducing new resolutions to the field of EDM machining.

The mechanical properties of hydrogen functionalized graphene (HFG) sheets werepredicted in this work by using artificial neural network approach. Thepredictions of tensile strength of HFG sheets made by the proposed approach arecompared to those generated by molecular dynamics simulations. The resultsindicate that our proposed computing technique can be used as a powerful toolfor predicting the tensile strength of the HFG sheet.

This paper facilitates realtime pursuit of an automobile and seeks to minimize the possibility of deaths by delay in the arrival of aid by alerting the concerned people about the mishap of the vehicle. According to a government survey, drowsiness and drunk driving constitute to 22 and 33 percent of accidents respectively in India. The number of lives lost can be diminished if the assistance can be procured at the earliest. To develop such a system which can notify the concerned people about the mishap, GPS module, GSM module, accelerometer is interfaced with Arduino uno which acts as the controller. The accelerometer detects the accident by a change in preset value of the vehicle orientation and sends the location through GPS module to registered sim card via GSM module without any indulgence of the driver or passengers. The planned system aims to cut back deaths in road accidents by quite nine p.c.

In an application such as literature recommendation, we require a comprehensive recommender model that can generate relevant recommendations similar to the literature provided in the input query. In this paper, we have proposed a novel content-based recommender system based on Latent Dirichlet Allocation (LDA) and Jensen-Shannon distance, which can be used specifically for the task of literature recommendations. We have compared this model with the standard cosine-similarity based approach for its use to generate scientific publication recommendations, in which recommend suitable journals/conferences to publish a research work based on the abstract of the user's manuscript as an input. We evaluated the results of both the proposed model and standard cosine-similarity based approach over unseen documents and achieved a precision score of 62.58% while the standard cosine-similarity based approach achieved a precision of only around 48%.

Due to present need of light weight and high performance structure, new hybrid materials are developed known as fiber metal laminates (FML). Fiber metal laminates comprises of alternate layers of composite and metals bonded together with appropriate adhesive technique. Mechanical and chemical surface treatment of 2024 aluminum alloy metal for obtaining strong interface bond between aluminum and composite layer. Due to decreased manufacturing cost and time, ease in forming, recyclability, thermoplastic composite are preferred in this study. Fiber metal laminates manufactured by press molding of one surface treated 2024 aluminum alloy as core material and two outer layers of glass fiber reinforced thermoplastic composite. This study investigates the mechanical properties of these fiber metal laminates by performing various tests such as tensile test, low velocity impact test, fire resistant test and flexural test. These results were further validated using analysis software.

The current virtualization technology induces an overhead in their performance. The benefits of Linux containers led to their development and increasing its usage among administrators and developers. Docker container offers an isolated space for running the application in the containerized environment. Containers are known for starting applications in quicker time as compared to those running on baremetal or virtualized environment. For calculating the startup time of containerized applications, we are using a benchmark based on Docker containers. This benchmark is capable of calculating the startup time of Docker application running services. The services can be Apache web server, Redis, PostgreSQL, and many others. The proposed system calculates the startup time for Docker containers running the Apache web server in baremetal, and virtual machine (KVM). The test will let the users know about the performance of Docker applications in both the environments. The analysis will help them in deciding the environment for working with Docker application to obtain maximum performance. The experimental setup of the proposed system shows that running containerized application on baremetal will improve the startup time than a virtual machine. The performance on baremetal is enhanced by about 50% than a virtual machine.

The primary function of a suspension system is to isolate the vehicle body from road irregularities thus providing the ride comfort and to support the vehicle and provide stability. The suspension system has to perform conflicting requirements; hence, a passive suspension system is replaced by the active suspension system which can supply force to the system. Active suspension supplies energy to respond dynamically and achieve relative motion between body and wheel and thus improves the performance of suspension system. This study presents modelling and control optimization of a nonlinear quarter car suspension system. A mathematical model of nonlinear quarter car is developed and simulated for control and optimization in Matlab/Simulink® environment. Class C road is selected as input road condition with the vehicle traveling at 80 kmph. Active control of the suspension system is achieved using FLC and PID control actions. Instead of guessing and or trial and error method, genetic algorithm (GA)-based optimization algorithm is implemented to tune PID parameters and FLC membership functions’ range and scaling factors. The optimization function is modeled as a multi-objective problem comprising of frequency weighted RMS seat acceleration, Vibration dose value (VDV), RMS suspension space, and RMS tyre deflection. ISO 2631-1 standard is adopted to assess the ride and health criterion. The nonlinear quarter model along with the controller is modeled and simulated and optimized in a Matlab/Simulink environment. It is observed that GA-optimized FLC gives better control as compared to PID and passive suspension system. Further simulations are validated on suspension system with seat and human model. Parameters under observation are frequency-weighted RMS head acceleration, VDV at the head, crest factor, and amplitude ratios at the head and upper torso (AR_h and AR_ut). Simulation results are presented in time and frequency domain. Simulation results show that GA-based FLC and PID controller gives better ride comfort and health criterion by reducing RMS head acceleration, VDV at the head, CF, and AR_h and AR_ut over passive suspension system.

The most widely and well known machining process used is turning. The turning process possesses higher complexity and uncertainty and therefore several empirical modelling techniques such as artificial neural networks, regression analysis, fuzzy logic and support vector machines have been used for predicting the performance of the process. This paper reviews the applications of empirical modelling techniques in modelling of turning process and unearths the vital issues related to it.

A small wind turbine blade was designed and optimized in this research paper. The blade plays an important role, because it is the most important part of the energy absorption system. Consequently, the blade has to be designed carefully to enable to absorb energy with its greatest efficiency. The main objective of this paper is to optimized blade number and selection of tip speed ratio corresponding to the solidity. The power performance of small horizontal axis wind turbines was simulated in detail using blade element momentum methods (BEM). In this paper for wind blade design various factors such as tip loss, hub loss, drag coefficient, and wake were considered. The design process includes the selection of the wind turbine type and the determination of the blade airfoil, twist angle distribution along the radius, and chord length distribution along the radius. A parametric study that will determine if the optimized values of blade twist angle and chord length create the most efficient blade geometry. The 3-bladed, 5-bladed and 7-bladed rotor achieved maximum values of Cp 0.46, 0.5 and 0.48 at the tip speed ratio 7, 5 and 4 respectively. It was observed that using BEM theory, maximum Cp varied with strongly solidity and weakly with the blade number. The studies showed that the power coefficient increases upto blade number B = 5, while the blade number if increased above 5 then the power coefficient decreases at operating pitch angle equal to 3°. Highest Cp would have solidity between 4% to 6% for number of blade 3 and design point tip speed ratio of about "7". Highest Cp would have solidity ranging from 5% to 10% for number of blade 5 and 7 and design point tip speed ratio of about 5 and 4 respectively.

The recent progress of the Internet of Things (IoT) for various smart city applications into the Industry 4.0 resolution offers different advantages and challenges. E-healthcare is one of the vital applications of Industry 4.0 emerging as the Healthcare 4.0 standard for remote health monitoring. Healthcare 4.0 is originally a sub-class of Industry 4.0 standard. The Healthcare 4.0 standard consists of different layers such as edge layer, fog layer, cloud storage layer, and blockchain layer. For E-healthcare systems, the main challenges are concerning medical data security and privacy-preserving while processing the data from edge-layer to cloud storage layer via fog computing. The development of blockchain technology connected with cloud storage and edge layer offers strong security provisions. However, there are not enough experimental works available to address lightweight cryptography with blockchain implementation. In this paper, the robust framework of securing the medical data processing using blockchain connected with the cloud storage system has proposed. The medical data collected at the edge layer first encrypted using Elliptic Curve Cryptography using Elliptic Curve Diffie Hellman (ECDH). The encrypted data has stored in cloud storage, and then it is reflected in the blockchain. For signature generation and authentication of medical data, the Elliptic Curve Digital Signature Algorithm (ECDSA) has been designed. The experimental outcome of the proposed framework outperforms the state-of-art solutions.

The integration of immunotoxicity data into chemical risk assessment paradigms is essential for comprehensively evaluating the potential health hazards posed by chemical exposures. This review provides a comprehensive overview of the methodologies, challenges, and future directions for integrating immunotoxicity data into risk assessment frameworks. It discusses the fundamental principles of immunotoxicology and its relevance to chemical risk assessment, highlighting the critical roles of the immune system in health and defense against harmful agents. Next, we explore traditional chemical risk assessment frameworks, including exposure assessment, hazard identification, dose–response assessment, and risk characterization, highlighting the need for incorporating immunotoxicity endpoints to enhance hazard characterization and risk estimation. Subsequently, we delve into dose–response modelling for immunotoxicity, elucidating principles, methods, and case studies illustrating dose–response relationships and extrapolation of data from animal studies to humans. Furthermore, we discuss hazard characterization of immunotoxicity, focusing on the identification of immunotoxic hazards, evaluation of immunotoxicity endpoints, and utilization of immunological biomarkers in risk assessment. We then examine cumulative risk assessment strategies, presenting a conceptual framework for assessing cumulative risks of immunotoxicity from multiple chemical exposures and methods for integrating exposure and hazard data from different chemicals. Lastly, we explore emerging trends and future directions in immunotoxicity risk assessment, including high-throughput screening assays, omics technologies, computational modelling, and alternative testing methods, along with potential regulatory implications and future research needs. This review provides valuable insights for researchers, regulators, and stakeholders involved in chemical risk assessment and public health protection, facilitating the development of evidence-based strategies for safeguarding human health from immunotoxic risks associated with chemical exposures.

The paper is based on the concept of Automation used in waste management system under the domain of Cleanliness and Hygiene. Dumping garbage onto the streets and in public areas is a common synopsis found in all developing countries and this mainly end up affecting the environment and creating several unhygienic conditions. In order to deal with these problems Smart netbin is an ideology put forward which is a combination of hardware and software technologies i.e. connecting Wi-Fi system to the normal dustbin in order to provide free internet facilities to the user for a particular period of time. The technology awards the user for keeping the surrounding clean and thus work hand in hand for the proper waste management in a locality. Smart netbin uses multiple technologies firstly the technology for measuring the amount of trash dumped secondly the movement of the waste and lastly sending necessary signals and connecting the user to the Wi-Fi system. The proposed system will function on client server model, a cause that will assure clean environment, good health, and pollution free society.

The custard apple is known as sugar-apple, sweetsop. It is known as sitafal in some state in India. Custard apple is high in energy and an excellent source of vitamin C and manganese. It also provides iron, phosphorus and potassium. This fruit has low cholesterol, sodium and saturated sodium, which is good for health. This paper gives a survey on leaf parameter analysis, detection of healthy, sick or affected region of the leaf and classification of leaf diseases by using different methods for various plant. It is crucial and difficult for human eyes to detect the exact type of leaf diseases by naked eyes. Each plant leaf has different symptoms of various diseases. The algorithm designed for one plant does not work accurately, with other plant's leaf. Specialized algorithms for the custard apple plant is required to detect leaf diseases along with the leaf parameter analyzer. To identify the custard apple leaf diseases accurately, image processing and machine learning techniques are helpful. Leaf area, length, width are some of the very important contributors in plant growth analysis and photosynthesis.

Environmental parameters like humidity, temperature, rainfall, wind flow, light intensity, soil pH are main factors for precision agriculture. Fluctuations in weather parameters like humidity, temperature and so on along with the inappropriate management result into a decrease in crop productivity. Therefore disease prediction is more important to beat these problems. The real-time update will alert the farmer by indicating which crop is in trouble, so the expenses on insecticides, pesticides will reduce and overall economic condition of farmers will improve. The proposed system gives more emphasis to predict diseases of the crop with the use of the Internet of Things and machine learning algorithms. Different sensors collect the real-time data of environmental parameters like temperature, humidity, rainfall, light intensity. Utilizing these data, crop diseases are predicted using machine learning algorithms. Such predictions would warn the farmers about crop diseases through text message or web browser. This work can be extended in the future to help farmers in other ways like which fertilizer can be used to overcome this disease problem.

In this study, the integrated Taguchi-simulated annealing (SA) approach is applied to examine the wear behaviour of silicon nitride (Si3 N4)-hexagonal boron nitride (hBN). Wear tests for Si3N4 -hBN composite versus steel (ASTM 316L) disc were carried out for a dry sliding conditions in a so-called pin-on-disc arrangement. The tests were realized at % volume of hBN 0, 4, 8, 12, 16 in Si3 N4 under the loads of 5, 10, 15, 20, 25 N. The wear rate (WR) was analyzed using Taguchi –signal to noise ratio approach with the aim of finding optimal combination of load and % volume of hBN in Si3N4 . By applying the analysis of variance, it was also found that the greatest impact on wear rate has interaction of load and % volume of hBN with percentage effect of 51.89%, then % volume of hBN with 35.04% and load with 13.06%. The experimental results are further utelized to develop the second-order, linear mathematical model. Further, this model is processed with simulated annealing (SA) to find the optimal combination of load and % volume of hBN to minimize wear rate. Combined Taguchi-SA approach was successfully used to predict the optimal combination of load and % volume of hBN in Si3 N4 to minimize wear rate of Si3 N4 . The dominant wear mechanism is adhesive wear as confirmed by scanning electron microscopy with energy dispersive spectroscopy (SEM-EDS).

Deepfakes allow for the automated gen- eration of fake video content, often accomplished through the use of generative adversarial networks. To address the increasing issue of deepfakes, this study focuses on constructing a model that incorporates advanced techniques. The researchers combined the ResNeXt, Long Short-Term Memory (LSTM), and ResNet architectures, selecting them based on their effectiveness in handling complex visual data and capturing temporal dependencies. Prior to detection, the dataset underwent pre-processing using the Multi-Task Cascaded Convolutional Neural Network (MTCNN), which facilitated the accurate extraction of facial regions. Importantly, the study evaluated the model across three diverse and significant datasets: the Face Forensics++ dataset (FF-DF), the Celeb-DF dataset, and the Facebook Deepfake Detection Challenge (DFDC) dataset. This comprehensive evaluation en- sured the model's ability to generalize and its suitability for real-world scenarios, as demonstrated by its exceptional detection accuracy. The combination of models employed in this study yielded highly accurate results and remained robust in the face of evolving deepfake technology.

Using the blockchain technology to store the privatedocuments of individuals will help make data more reliable and secure, preventing the loss of data and unauthorized access. The Consensus algorithm along with the hash algorithms maintains the integrity of data simultaneously providing authentication and authorization. The paper incorporates the block chain and the Identity Based Encryption management concept. The Identity based Management system allows the encryption of the user's data as well as their identity and thus preventing them from Identity theft and fraud. These two technologies combined will result in a more secure way of storing the data and protecting the privacy of the user.