MACHINE LEARNING-DRIVEN STATISTICAL ANALYSIS OF INDIAN RESTAURANTS: INSIGHTS FROM THE ZOMATO DATASET Ayushi Vaidhy, Deepak Batham, Rachit Jain, Amit Kumar Manjhvar Facta Universitatis Series Electronics and Energetics, 2025 Advances in technology and web applications, such as Zomato, have significantly transformed the restaurant industry by catering to diverse culinary preferences and offering a wide variety of food options to customers. This platform stores a vast amount of data that can be analyzed for valuable insights. The paper examines dining habits and restaurant performance through exploratory data analysis (EDA) and machine learning (ML) algorithms, helping customers find the best restaurants based on cost, ratings, location, food quality, and service. The study applies several ML models, including Linear Regression (LR), Decision Tree (DT), Random Forest (RF), Gradient Boosting (GB), XGBoost, K-Nearest Neighbors (KNN), and LASSO to the Zomato dataset. The results are evaluated using metrics such as accuracy, mean absolute error (MAE), model fit time, and model prediction time. Among these models, DT and RF show the highest predictive accuracy, with RF achieving 97.86% and outperforming other algorithms. These findings provide restaurant owners with valuable insights to enhance customer satisfaction, optimize pricing, and improve service quality. The study also demonstrates the important role of ML in the restaurant industry and suggests future opportunities for integrating realtime data, deep learning models, and sentiment analysis to offer even more precise predictions and insights.
Prediction of Benign and Malignant Breast Cancer using Machine Learning Techniques: A Review and Comparative Analysis Amit Manjhwar, Deepak Batham, Mahesh Parmar, Vikas Sejwar 2024 1st International Conference on Advanced Computing and Emerging Technologies Acet 2024, 2024 This Breast cancer (BC) is a predominant illness majorly found in women which may cause death if predicted in an uncontrolled stage. In BC, breast cell rises uncontrolled and infects the surrounding part of the breast. To detect BC in premature stages using simple, rapid, and correctly detected methods, machine learning (ML) techniques played a vital role. This paper mainly emphasizes different ML algorithms such as Classification and Regression Tree (CART), K-Nearest Neighbor (KNN), Random Forest (RF), Support Vector Machine (SVM), and Naive Bayesian (NB) to detect benign and malignant BC then a comparative analysis of these algorithms has been performed and tested on the standard BC datasets. These datasets were pre-processed first to minimize error after that ML models were trained and tested. The investigated experimental results of ML algorithms are evaluated on accuracy, recall, precision, and F1 score metrics. The SVM technique shows 98.83% accuracy and significantly improved performances on other metrics. These discoveries could help to the early prediction of BC and save life.
Public Sentiment Assessment of Coronavirus-Specific Tweets using a Transformer-based BERT Classifier Kanak Mahor, Amit Kumar Manjhvar International Conference on Edge Computing and Applications Icecaa 2022 Proceedings, 2022 Worldwide, the (COVID-19) pandemic had also affected people's daily routines. In general also during lockdown periods, people around the world use social media to express their thoughts and feelings about the epidemic which has interrupted their daily lives. There has been a huge spike in tweets about coronavirus on Twitter in a short period of time, including both positive and negative messages. As a result of the wide range of content in the tweets, the researchers have turned to sentiment analysis in order to gauge how the general public feels about COVID-19. According to the findings of this study, the best way to examine COVID-19 is to look at how people use Twitter to share their thoughts and opinions. Sentiment categorization can be accomplished by utilising a variety of feature sets as well as classifiers in combination with the suggested approach. Tweets collected from people with COVID-19 perceptions can be used to better understand and manage the epidemic. Positive, negative, as well as neutral emotion classifications are being used to classify tweets. In this study, Tweets containing specific information about the Coronavirus epidemic are used as sentiment analysis packages. Bidirectional Encoder Representations from Transformers (BERT) are used to identify sentiment categories, whereas the TF-IDF (term frequency-inverse document frequency) prototype is used to summarise the topics of postings. Trend analysis and qualitative methods are being used to identify negative sentiment traits. In general, when it comes to sentiment classification, the fine-tuned BERT is very accurate. In addition, the COVID-19-related post features of TF-IDF themes are accurately conveyed. Coronavirus tweet sentiments are analysed using a BERT and TF-IDF hybrid classifier. Single-sentence classification is transformed into pair-sentence classification, which solves BERT's performance issue in text classification problems. Our evaluation measures (accuracy= 0.70; precision= 0.67; recall= 0.64; and F1-score= 0.65) are used to evaluate the effectiveness of the classifier.
Relation classification from unstructured medical text using feature based machine learning approach Saumaya Gupta, Amit Kumar Manjhvar Proceedings International Conference on Trends in Electronics and Informatics Icei 2017, 2017 There is a lot of useful information available in medical documents. Information as medical named entities, relationship between medical entities, medical summary and etc. Most of the time such information in medical documents is unstructured and available in nonstandard natural language so it is difficult to automatically collect and present this information in a structured way. Structured information can be present as clinical entity in the text, relationship between clinical entities, summary of the text, etc. To get the specific information from the text, many rule based and machine learning techniques are widely used. In this article we propose a feature based machine learning model for relation classification task. We will also discuss a relative comparison with existing relation classification model.
A model for forecasting dengue disease using genetic based weighted FP-growth Vandana Rajput, Amit Manjhvar Proceedings International Conference on Trends in Electronics and Informatics Icei 2017, 2017 In few years, Data Mining is a big and motivating research area in medical & healthcare department. It is helpful to find profitable and successful systematic method in wellbeing data. It is helpful to forecast the different diseases like-Dengue fever, Cancer, Diabetes, heart disease etc. A big agreement of reading former conceded out on disease detection using an optimization technique to palliate the drawbacks of conventional approaches. In our paper, we have to design a novel model for forecast the dengue disease. Here, we use genetic algorithm to calculate the actual weight of attributes afterwards applied the FP-Growth with actual weight. Theoretical study and experiments have displayed that the modified approach is able to detect the virtual significance of attributes in requisites of their weights. This model are deliberate and the parameters are set to get optimal forcast performance. At last, the outcome displays that the model produces the better prediction.
An improved optimized clustering technique for crime detection Nidhi Tomar, Amit Kumar Manjhvar 2016 Symposium on Colossal Data Analysis and Networking Cdan 2016, 2016 Data mining automates the finding predictive records procedure in big databases. Clustering is a most famous method in data mining and is an important methodology that is performed based on the similarity principle. The segregation of a big database is a stimulating and task of time consuming. It concludes two different stages: first, feature extraction maps all documents or record to a point in the space of high-dimensional, then algorithms for clustering automatically grouping the points into a cluster hierarchy. Clustering has various applications in different fields. Few of the fields include criminology, text mining, image resolution, machine learning. Crime detection has become one of the most attractive field as the crime rate in India and whole world is increasing at a greater pace. We as citizens of a country have to contribute towards its detection and removal. Thus, a comprehensive survey carried about the basics of clustering has given in this paper. Moreover, proposed work was given that gives the idea of the work going to be done in the upcoming time.
A review on link prediction in social network Ajay Kumar Singh Kushwah, Amit Kumar Manjhvar International Journal of Grid and Distributed Computing, 2016 Social network analysis is an evolving field of research and link prediction problem shows a vital role for prediction of social network structure. This paper emphases on prevailing research on link prediction problem. Prevailing researches reveal that link prediction problem complexity, available solutions effective group communication management and social link consciousness. The link prediction problem across associated networks can include anchor link prediction problem and link transfer through associated heterogeneous networks. This paper summarizes recent growth about link prediction algorithms and survey of all the prevailing link prediction techniques.
