Journal of Big Data Analytics and Business Intelligence

Smart Sanitation System: An IoT and AI-driven Approach for Proactive Odor Management

2026-02-26T11:43:31+00:00

Conventional urban sanitation management is inherently inefficient, such that an existing sanitation management process is only based on fixed urban cleaning schedules without any consideration of constantly varying usage dynamics. As a consequence of these inefficiencies, public facilities often end up being degraded due to the quick formation of malodors coupled with equally poor hygiene standards. This study discusses a novel smart sanitation system that provides the possibility of easily evolving from such conventional inefficiencies to higher efficiencies. In achieving this objective, the system architecture relies on an IoT sensor network comprising ESP32 microcontrollers with integrated MQ135 gas sensors and DHT11. The data collected is then processed to predict odor surges hours before they are detectable to human detection. With such a prediction, targeted condition-based maintenance alerts will be sent to cleaning personnel to ensure they are utilized only when necessary. The experimental results provided in this research demonstrate the predictive accuracy of this technique, which significantly reduces wastage of resources while maximizing efficiency. Additionally, this technique can help provide large-scale national programs like Swachh Bharat Abhiyan with improved user comfort and dignity. Ultimately, this research presents an excellent tool to incorporate intelligent sanitation management into the overall environment of smart technology.

Smart Contractual Risk Evaluation System Using Machine Learning

2026-03-12T09:49:34+00:00

This study presents a systematic literature review and proposes a conceptual framework for smart contractual risk evaluation systems using artificial intelligence (AI) and machine learning (ML), without conducting empirical implementation or experimental validation. AI models, particularly those based on natural language processing (NLP) have shown promising results in automating contract analysis and risk assessment. This review examines key AI-based techniques, including LegalT5, LawGPT, and K-means clustering, in enhancing contract summarisation, clause extraction, and risk evaluation. By leveraging deep learning models fine-tuned on domain-specific legal corpora, these systems enable efficient clause identification, contextual understanding, and semantic risk categorization. Furthermore, the integration of abstractive summarization models with unsupervised clustering algorithms enhances interpretability by grouping similar contractual clauses and identifying potential areas of risk or non-compliance. The paper also examines hybrid AI techniques such as expert systems, artificial neural networks (ANN), and adaptive neuro fuzzy inference systems (ANFIS) that support decision-making under uncertainty. Overall, this review provides a comprehensive overview of current AI-driven methodologies in contract risk analysis and outlines future directions for developing explainable, scalable, and legally compliant intelligent systems.

A Smart AI-based Conversational System for Enhancing Disease Awareness

2026-03-28T08:40:56+00:00

Chatbots have emerged as scalable tools in healthcare and public health for tasks such as triage, risk assessment, information delivery, behaviour-change coaching, and mental-health support. Rapid adoption during the COVID-19 pandemic showcased their value in managing high information demand and reducing clinical workloads. Recent reviews identified promising outcomes of chatbots in certain applications, notably smoking cessation, such that results are inconsistent with diet and physical-activity interventions. Positive user engagement and feasibility have been reported by many systems, although methodological limitations and short follow-up periods constrain conclusions about long-term effectiveness. Several studies also highlight the role of continuous monitoring and user feedback in improving system performance and personalisation over time. Furthermore, integrating human oversight with AI-driven systems helps ensure reliability and supports more informed decision-making in real-world applications. Ethical considerations, including privacy, accuracy, equity, and transparency, continue to inform deployment. This literature survey synthesises current applications, findings on effectiveness, and research gaps toward guiding safer use of evidence-based chatbots across diverse healthcare and public-health settings worldwide today.

Transforming Business Decisions through Artificial Intelligence: A Comprehensive Study

2026-03-28T08:59:32+00:00

The integration of artificial intelligence (AI) into business decision-making processes has emerged as one of the most consequential technological developments of the contemporary era. This paper examines the multifaceted impact of AI on organisational decision-making, encompassing strategic, operational, financial, and human resource management domains. Drawing upon existing scholarly literature and industry case studies, the study investigates how AI technologies such as machine learning, natural language processing, predictive analytics, and robotic process automation are reshaping how business leaders formulate, evaluate, and execute decisions. The research explores the economic implications of AI adoption, including productivity gains and cost reduction. Critical challenges such as algorithmic bias, ethical accountability, data security vulnerabilities, and employee displacement are also analysed. The study reveals that organisations that strategically align AI adoption with core business objectives demonstrate superior decision-making outcomes. Findings suggest that AI is not merely a technological upgrade but a transformative force necessitating a fundamental reimagining of organisational structures, leadership competencies, and decision-making philosophies. The paper concludes by identifying future directions with emphasis on explainable AI, human–AI collaboration frameworks, and governance mechanisms.

An Efficient Content-Based Movie Recommendation System Using TF-IDF and Cosine Similarity: Design, Implementation, and Evaluation

2026-03-28T09:19:03+00:00

With the rapid growth of online streaming platforms, users are often overwhelmed by the large number of available movies. Identifying relevant content based on individual preferences has become a challenging task. This paper presents the design and implementation of a content-based movie recommendation system using Python. The system analyzes movie metadata such as genre, cast, keywords, and description to generate personalized recommendations. TF-IDF vectorization is applied to convert textual data into numerical feature vectors, and cosine similarity is used to measure similarity between movies. Based on these similarity scores, the system recommends movies that closely match user preferences. To enhance the user experience, additional features such as trailer access, a watchlist, and a favourites list are integrated into the system. An interactive user interface is developed using Streamlit, allowing users to explore recommendations in real time. The proposed system is efficient, scalable, and suitable for both academic and practical applications, providing accurate recommendations along with improved user engagement.