3830 research outputs found
Sort by
Comparative analysis of clustering algorithms for customer segmentation and improved marketing strategies
No full textThis study evaluates the effectiveness of K-Means, DBSCAN, and OPTICS clustering algorithms for customer segmentation. Using the Recency, Frequency, and Monetary (RFM) model, customer value was quantified, and customers were segment based on their transactional behavior.
Dataset was obtained from UCI machine learning repository and contained transaction details of an online retail business. The data underwent cleaning, feature engineering, normalization, and dimensionality reduction using UMAP. The clustering algorithms were then applied and evaluated using Silhouette Scores and Davies-Bouldin Indices.
K-Means effectively grouped customers, achieving a Silhouette Score of 0.445 and a Davies Bouldin Index of 0.736. DBSCAN handled noise and identified arbitrary shapes but produced scattered clusters with a lower Silhouette Score of 0.132 and a higher Davies-Bouldin Index of 1.435. Although OPTICS had similar scores to DBSCAN, it resulted in smoother clusters and handled varying densities more effectively than DBSCAN.
To summarize, K-Means provided the best cluster separation. DBSCAN and OPTICS were better for noise handling and variable densities
Analyzing CO2 Emission Intensity: A Comprehensive Study of Clean and Unclean Energy Sources using ML Techniques
No full textThis thesis presents a comprehensive analysis of CO2 emission intensity, examining the intricate interplay of energy sources and their contributions to greenhouse gas emissions. The study evaluates the impact of various energy sources, categorizing them into two main groups: clean energy (wind, solar, hydro, bioenergy, and nuclear) and unclean energy (coal, gas, and other fossil fuels). By
utilizing Generalized Linear Models (GLM), this research offers a robust prediction of CO2 emission intensity, providing insights into the relative contributions of different energy sources on a regional and global scale. This analysis, which includes the percentage of energy usage from each source, allows for a more accurate quantification of CO2 intensity, simplifying th e process of rebalancing energy dependence to reduce environmental impact. Furthermore, this research employs Time Series Forecasting Techniques, specifically the AutoRegressive Integrated Moving Average (ARIMA) model, to forecast the trends in CO2 intensity across various regions. These forecasting methods facilitate a deeper understanding of how CO2 emissions are expected to evolve over time and allow for the identification of critical points for intervention and mitigation strategies. Findings reveal significant variations in CO2 emission intensity across energy sources and regions, shedding light on the key players in our environmental challenges. The study's data -driven analysis, incorporating energy usage percentages, offers insights into the relative contributions of different energy sources to CO2 emission intensity and underscores the critical importance of transitioning toward cleaner, more sustainable energy alternatives. This research serves as a valuable resource for policymakers, energy industry stakeholders, and environmental advocates, providing empirical guidance for mitigating the environmental impact of energy production and offering a quantifiable basis for rebalancing energy dependence
Examining the impact of project based learning on secondary grade teachers' practices and professional development
Full text linkIn these running lives the study of the significant impact of Project-Based Learning (PBL) on secondary education, with the principal center of attention is the way it alters the teaching approaches and the way the teachers are grown professionally. The qualitative research of this study discovers that PBL builds an online platform, which turns studying into the episode less focused on pupil and the one that includes more learner and teacher interaction. The role of teachers switches from class leaders to facilitators that in turn set up the critical thinking and collaboration atmosphere for students. The application of PBL inevitably increases students' involvement and involvement in the education process and it provides educators with proof of the utility of innovative methods in the process of learning which encourages them to continue using such techniques in their lessons. The document reminds that efficient resources taken together with implementation of administrative support to the PBL are needed for the purpose of integrating this teaching approach into the existing curricula and stresses that this teaching method must be adjusted to the specific educational contexts. This study furnishes us meaningful evidence towards increasing the scale of power of PBL and its long term effect, suggesting PBL as a revolutionary future of education
Assessing the effectiveness of Green Supply Chain Management in Enhancing Sustainability in the UK retail industry: A focused study on Walmart Inc.
No full textThe business's sustainable growth policy is crucial in attracting loyal customers. Providing sustainable items to customers is a strategic step that may boost the business's financial performance and sales volume. Customers' shifting requirements in the retail sector encourage businesses to consider the practical implications of green supply chain management in the global supply chain management process to offer significant product items to customers. Increasing knowledge regarding Environmental Protection is a measure to change the consumers' demand in this sector. As a result, providing and sustaining green supply chain operations became the fundamental factor for the valuable integration of a sustainable product development strategy within Walmart, which helped both the surrounding environment and the company's profitability due to the effective increment in the sales volume.
The study aims to evaluate the impact of Walmart's green supply chain management (GSCM) methods on sustainability performance. The report evaluates Walmart's systematic approach to GSCM connected to waste and produc tion adjustments as a present development. The research also attempts to investigate the impact of Walmart's GSCM policies on waste creation and energy efficiency, hence enhancing the Company's operating strategy. The research assesses Walmart's shift from customer focused activities to supplier centric initiatives by adjusting its ecological footprint without raising product pricing. Also to support the progress and investigate GSCM methods that might be implemented to improve Walmart's sustainability performance
Analyzing annual report financial data to understand its impact on share prices of Indian IT consulting firms using machine learning algorithms
No full textInvestors, financial analysts and individuals or anyone for that matter can rely on Annual Reports to decide on the firm’s current performance and to understand its business to make calculated plans on their investments. Financial information inside annual reports provides an overview of past performance, how they make use of the external environment for their growth needs, their strategies for growth and their expectations for the future. This research tries to investigate how financial data from annual reports along with market data can be used by ML algorithms like Random Forest (RF), Gradient Boosting Machines (GBM), eXtreme Gradient Boosting (XGB) and CatBoost (CB) to identify most impactful financial features in process of predicting firm’s share price using feature selection techniques like P-IMP, FFS, BFS and RFE. The results suggest that XGB performs the best for this financial dataset of IT consulting firms overall, with a low MAE and MSE score across K-fold validation capturing the variance of data with a high R2 score with less susceptibility to outliers and also being more computationally efficient compared to others. The feature selection methods of P-IMP, FFS, BFS and RFE all concur in outputs highlighting the market data features as most influencing features in predicting share prices followed by balance sheet data features providing an insight into focusing on given list of features. Thus, data present in annual reports can be effectively analyzed using ML algorithms by an individual to make their investment decisions based on firm’s actual performance rather than speculation
Water Quality Analysis Using Machine Learning
No full textWater quality evaluation is crucial in environmental management, and utilising machine learning models improves the accuracy of predictions. This study aims to compare different machine learning models for predicting water quality before and after the monsoon season in Telangana. The dataset used in this research was obtained from the Telangana Ground Water Department. The chosen models, namely Random Forest Classifier (RFC), Support Vector Classifier (SVC), Multi-layer Perceptron (MLP), Stochastic Gradient Descent (SGD), and KNeighborsClassifier, are assessed with a specific focus on imbalanced data using Principal
Component Analysis (PCA) as the model was giving perfect score due to being imbalanced which was incomparable and incorrect. The effectiveness of the models is evaluated by employing essential performance metrics, including recall, precision, and F1 score as the accuracy does not work well with imbalanced data.
The pre-monsoon results indicate that RFC performs exceptionally well, with a recall of 0.988 and precision of 0.900. The monsoon transition has had a noticeable effect on RFC, as it continues to perform exceptionally well in the post-monsoon period, with an improved recall rate of 0.996 and precision rate of 0.993. SVC, SGD and MLP demonstrate consistent and strong performance in both time periods, demonstrating their ability to adapt. Notably, the KNeighborsClassifier demonstrates enhancement after the monsoon season, highlighting its sensitivity to seasonal changes.
The analysis of seasonal variations was performed with help of T-test on the machine learning model performance’s. RFC demonstrates consistent excellence. The comparative analysis enhances the scientific comprehension of machine learning models in predicting water quality, providing practical implications for environmental scientists, policymakers, and stakeholders involved in water resource management
Prediction of Clients of “The Career Coach Company” based on Customer Profiling: Leveraging Machine Learning for Marketing Strategies Optimization
No full textThis research aims to predict the intention of a client to buy the services of The Career Coach Company by implementing machine learning classification algorithms such Random Forest Classifier and K Nearest Neighbors, based on the customer profile of the actual clients and the people who got engaged with the Company at some point. Data visualisations were generated to help the Company comprehend the big picture of the most significant variables related to the
current clients. This proposal presents the research plan for establishing a predictive model that can provide insightful information based on customer attributes helping to boost its marketing strategies and optimize revenue generation. By employing analytical approaches, the study strives to contribute to the growing field of business analytics by offering useful insights for the Company's business growth and market competitiveness, helping it to address its marketing campaigns based on the actual profile of its clients
Text Analysis Of Russia-Ukraine War
No full textIn the Research, we assess the use of Twitter tweets to identify Fake News. Knowing the intent of the text can help classify the News as fake or not, and it can help in many policy makings and understanding the public opinion on any given topic. The study's primary purpose is to identify fake News based on tweets from the Russian and Ukrainian wars. The following is an ongoing war, termed cyber war, because of the use of social media. The study uses data from Twitter by writing the Python script and extracting the code in an HTML document with the help of understanding natural language and converting Python text analytics into a raw structured format. The Research seeks to find the label based on the data's subjectivity and polarity.
Given that, clean text data is converted into vectors with the help of the TF-IDF metric. The values are fed into various machine-learning algorithms and tested on different accuracy matrices. The random forest model has achieved the maximum accuracy of 87% in identifying fake and real tweets. The given model can help identify fake News on the Internet and help reduce phoney content. The provided search has used a standard data mining approach concerning text analytics based on the given topic
A Web Application for a Call Center's Technical Support Department
No full textThis project is aimed to revolutionize how an internet provider handles the technical issues reported by their customers. Typically, when an internet connection is lost, intermittent or slow, the customer needs to ring the technical department for assistance, waiting in virtual queues to be connected to an available agent. The agent is generally highly stressed, as this is a highly stressful and repetitive job. It involves continuously appearing to be content, professional and calm while speaking repetitively to irate customers, following repetitive scripts and ensuring the calls are kept to as minimal time as possible. This project provides a solution, it is a Web Application that prompts users to select their issue and input the data required to troubleshoot. The software prompts and gathers required information from the user, the same questions and answers that the call centre agent would have gathered over the phone. The project results have been highly promising. With efficiently functioning forms in place to gather the required data and optional user registration, it is promising to be deployed and highly effective in a professional setting
Blockchain-Enabled Peer-to-Peer Energy Trade in Smart Homes in India
No full textThis research examines the decentralised peer-to-peer energy trade model within the residential household grid in India using blockchain technology and solar energy generation techniques. It addresses the gaps in existing literature and examines the viability of fixed transaction prices, role of blockchain technology in securing these transactions and validating the feasibility of the model in the households of Uttar Pradesh. By leveraging the empirical data, the research highlights the economic opportunity for both the consumers and the prosumers of the energy trade. It also emphasizes the promotion of adoption of renewable sources for energy generation while resolving the energy theft and improving the overall efficiency of performing the transactions using the encryption aspect of the blockchain technology. The model aids not only in securing the transactions taking place over the grid but also presents a significant measure that should be undertaken towards sustainability. Ultimately, the research shed light onto the potential of the proposed system for revolutionizing the energy market in India, promoting technological innovation and encouraging environmentally conscious practices