Proceedings of the 7th International Conference on Communications and Cyber Physical Engineering

Table of Contents

1. A Novel Approach of Superstore Sales Data by EDA and ARIMA

   J. Bhoomika Reddy, N. Ashritha, Katiki Reddy Navya, V. Kakulapati, V. Monica

   This chapter delves into the critical role of accurate sales forecasting in retail, particularly for superstores. It highlights the importance of understanding product demand and consumer behavior to optimize inventory and operational efficiency. The study focuses on analyzing superstore sales data from 2014 to 2021, emphasizing the impact of external factors like social media and the COVID-19 pandemic on market trends. The research employs Exploratory Data Analysis (EDA) and the ARIMA model to forecast future sales, providing valuable insights into trends, seasonality, and anomalies. The methodology includes data collection, preprocessing, EDA, model selection, and evaluation using metrics like MAE, MSE, RMSE, and MAPE. The study concludes that accurate forecasting can significantly enhance business decision-making, inventory management, and overall operational efficiency. The findings underscore the importance of periodic model review and adjustment to maintain accuracy. Future enhancements could involve incorporating external variables and advanced computational techniques to further refine sales forecasting.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

2. AI-Enabled Facial Redesign: Crafting Personal Features with Generative Adversarial Networks

   Sunil Bhutada, V. Kakulapati, K. Goutham, P. Nandith, Gulab Singh

   This chapter delves into the groundbreaking advancements in AI-enabled facial redesign, focusing on the use of Generative Adversarial Networks (GANs) to craft personalized facial features with remarkable realism. The text explores the technical aspects of GANs, including the dual-attention mechanism and the use of controlled GANs to address challenges such as posture and lighting disparities. It also examines the potential applications of AI-enabled facial redesign in various domains, from cosmetic surgery simulations to forensic science. The chapter highlights the ethical considerations and privacy concerns associated with this technology, emphasizing the importance of responsible development and regulation. Additionally, it reviews existing works in the field, providing a comprehensive overview of the current state of research. The proposed methodology outlines the steps for implementing AI-enabled facial redesign, including data collection, model selection, and evaluation metrics. The chapter concludes with a discussion on the future enhancements and the potential societal impacts of this technology, offering a holistic perspective on the transformative potential of AI-enabled facial redesign.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

3. A Novel Approach for Recognizing and Eliminating Escalation Attack Using AI Techniques

   Banoth Suman, Panga Saikiran, Nagelli Kasim, V. Kakulapati, M. Swapana Kamakshi

   This chapter explores a novel approach for recognizing and eliminating escalation attacks using AI techniques, focusing on the security challenges in cloud computing environments. The study evaluates the performance of various machine learning algorithms, including Random Forest, AdaBoost, XGBoost, and LightGBM, in accurately classifying insider threats. The methodology involves data collection from CERT databases, preprocessing, feature selection, and the application of machine learning techniques. The results highlight the superior performance of the voting classifier, which combines predictions from multiple models, achieving the highest accuracy of 96.45%. The study also discusses the importance of data processing, feature selection, and the optimization of model parameters to enhance the effectiveness of threat detection. The findings suggest that the proposed approach can significantly improve the precision of identifying insider threats, providing a robust solution for enhancing cloud security.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

4. Risk Analysis of Mental Health Using Chatbot Based on Text Detection Model

   M. Nagaraju, V. Kakulapati, P. V. Vaishnavi, A. Neha, M. Vaishnavi

   This chapter delves into the transformative potential of chatbots in mental health care, focusing on risk analysis through advanced text detection models. It explores the global impact of mental health issues, the shortage of mental health professionals, and the role of AI in bridging this gap. The study highlights the use of Natural Language Processing (NLP) techniques and various machine learning algorithms like LSTM, Random Forest, SVM, and Neural Networks to analyze textual data from chatbot interactions. The methodology section provides a detailed overview of the development process, including the use of Python and Django for implementation. The discussion section addresses the ethical and technological challenges of using AI in mental health care, emphasizing the need for further research. The conclusion underscores the potential of these AI-driven systems to provide early diagnosis, tailored treatment, and continuous support, paving the way for more accessible and effective mental health care.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

5. Unveiling the Potential of Deep Learning in Stock Market Forecasting: A Comparative Analysis

   K. Sneha Reddy, Tanusha Meka, Hema Sreyalahari Karanam, Pallavi Nooka

   This chapter delves into the potential of deep learning models, particularly Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks, in forecasting stock market trends. It begins with an overview of stock analysis techniques, including fundamental and technical analysis, and the importance of key indicators in understanding market dynamics. The chapter then provides a detailed explanation of LSTM and GRU architectures, highlighting their unique features and mathematical models. Data pre-processing techniques, such as MinMax scaling, sequence creation, and data reshaping, are discussed to prepare the stock data for analysis. The chapter presents a comprehensive analysis of stock data from major companies like Google, Microsoft, Apple, and Amazon, using various visualizations such as histograms, box plots, scatter plots, and line graphs. It also explores the relationship between risk and expected returns, as well as data trends and seasonality. The core of the chapter focuses on stock data prediction using LSTM and GRU models, comparing their performance through various metrics. The results indicate that the GRU model outperforms the LSTM model in terms of evaluation metrics, making it a preferred choice for stock market forecasting. The chapter concludes by discussing the factors to consider when choosing between GRU and LSTM architectures, emphasizing the importance of processing capacity, interpretability, and predictive accuracy. This detailed comparison provides valuable insights into the effectiveness of deep learning models in stock market forecasting, making it a crucial read for professionals seeking to leverage these technologies for better investment decisions.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

6. Design of Immersive Multi-level Parking

   C. Harinatha Reddy, Y. V. Siva Reddy, A. Pradeep Kumar Yadav, T. Bramhananda Reddy, N. Ravi Sankara Reddy, G. Raghu Ram

   This chapter delves into the design of an immersive multi-level parking system, addressing the critical issue of space optimization in urban areas. The text highlights the use of advanced technologies such as Building Information Modeling (BIM), Virtual Reality (VR), and Twin Motion to create a realistic and efficient parking solution. It provides a detailed analysis of the structural elements, including slabs, beams, columns, and footings, and their design considerations. The chapter also discusses the methodology used for implementing the idea, including the use of Total Station for precise measurements and AutoCAD for detailed planning. Additionally, it showcases the proposed parking structure for GPRE College, demonstrating how existing parking areas can be optimally utilized to accommodate both two-wheelers and four-wheelers. The conclusion emphasizes the increased efficiency in space utilization and the aesthetic improvements brought about by the new system.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

7. PV and Wind Integration of Microgrid Protection Scheme in Using Wavelet and Machine Learning

   Ravi Kumar Goli, Ramakrishna Ganji, Madhulatha Bethala, Vijay Chukka, Gopi Chand Govathoti

   This chapter explores the integration of solar and wind energy sources into microgrids, focusing on creating an efficient protection scheme. The study employs wavelet analysis and machine learning techniques to address various fault scenarios, ensuring the microgrid's safety and reliability. Key topics include the use of the biorthogonal 1.5 wavelet for signal analysis, the application of machine learning algorithms for fault detection and classification, and the examination of different fault types such as single line-to-ground and double line-to-ground faults. The research demonstrates how the proposed scheme effectively identifies, classifies, and locates faults in multiple zones, enhancing the overall stability and dependability of the power grid. The conclusion highlights the potential of the proposed method to improve microgrid performance and suggests future work to refine the algorithms and explore advanced wavelet functions.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

8. Development of GPS Controlled Solar Battery-Operated Self-navigating Vehicle

   G. Prasad Acharya, A. Gnana Priya, R. Sreema Reddy, K. Charitha

   This chapter explores the development of a GPS-controlled, solar battery-operated, self-navigating vehicle prototype. The system utilizes an Arduino development board and GPS module for autonomous navigation, with waypoints defining the path. Ultrasonic sensors ensure obstacle detection, enhancing safety. The vehicle is powered by solar energy, stored in a battery, making it eco-friendly. The text delves into the system architecture, workflow, and circuit implementation, providing a comprehensive overview of the prototype's design and functionality. The prototype's ability to navigate autonomously and its integration of renewable energy sources make it a notable advancement in autonomous vehicle technology. The chapter concludes with potential upgrades, such as integrating AI for full autonomy, highlighting the system's scalability and future applications.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

9. Automatic Motorcyclist Helmet Rule Violation Detection Using LIME or RNN and PNN

   Sunil Bhutada, V. Kakulapati, Kariveda Pravalika Reddy, Gurram Dhanu Sree, Mantha Aditi

   This chapter explores the development of an advanced helmet detection system using machine learning techniques, specifically focusing on Probabilistic Neural Networks (PNN) and Recurrent Neural Networks (RNN). The research aims to address the critical issue of helmet non-compliance, which contributes to a significant number of fatalities and injuries in traffic accidents. The methodology involves a comprehensive data collection process, exploratory data analysis, and the implementation of various machine learning algorithms. The system's performance is evaluated using metrics such as accuracy, precision, recall, and F1-score, ensuring its reliability and effectiveness in real-world scenarios. The integration of ensemble learning methods enhances the system's ability to capture intricate features and sequential patterns relevant to helmet detection. The article also discusses the potential impact of the system on safety compliance and accident prevention, highlighting its transformative potential in ensuring public safety and well-being. The conclusion emphasizes the importance of leveraging advanced machine learning methodologies to promote safer environments and the future enhancements that could further elevate the system's effectiveness and versatility.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

10. Generating Synthetic Images from Text Using RNN and BiLSTM

    Subhani Shaik, V. Kakulapati, Gudur Sathwik Reddy, Thumma Manoj, T. Bhargav

    This chapter explores the innovative approach of generating synthetic images from textual descriptions using a combination of Recurrent Neural Networks (RNN) and Convolutional Neural Networks (CNN). The study delves into the architecture that integrates RNNs for processing sequential text data and CNNs for extracting image features, aiming to create visually coherent images that correspond to given textual descriptions. The methodology involves training the model on the Flickr Text and Image dataset, with a focus on data preprocessing, model training, and text-to-image generation. The implementation analysis highlights the effectiveness of the CNN-BiLSTM algorithm, achieving an accuracy of 68%. The discussion section addresses the challenges in text-to-image synthesis and suggests future enhancements, such as training on diverse datasets and incorporating multimodal models. The conclusion emphasizes the promising results of using RNN and CNN for synthetic image generation and its potential applications in various fields.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

11. Human Body and Cloth Segmentation

    Nagaratna P. Hegde, Sireesha Vikkurty, Sriperambuduri Vinay Kumar, Chintaboina Mallikarjun, Mudavath Revanth

    This chapter delves into the world of human body and cloth segmentation, a critical area of research with wide-ranging applications. The text begins with an introduction to semantic segmentation and human parsing, highlighting their importance in various fields such as medical imaging, pedestrian detection, and self-driving cars. It then explores the U2Net architecture, a state-of-the-art deep learning model used for semantic segmentation. The chapter provides a detailed description of the LIP dataset, which is used to train and evaluate the model. The preprocessing techniques, including image resizing, normalization, and data augmentation, are discussed in detail. The text also covers the encoder-decoder structure of U2Net, with a focus on the DenseNet backend and the pyramid pooling module. The training and testing process is thoroughly explained, along with the evaluation parameters used to assess the model's performance. The results and discussion section provides insights into the model's accuracy and robustness. The chapter concludes with a look at the potential applications and future directions in the field of human body and cloth segmentation.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

12. Cyber Crime Detection Using Machine Learning

    M. Jyothirmai, M. Jayalakshmi, C. Ahalya, L. Lakshmi Prasanna Kumar

    This chapter delves into the critical issue of cybercrime detection, highlighting the increasing threat posed by cybercriminals to individuals, businesses, and governments. It explores the evolution of cybercrime, from early isolated incidents to sophisticated, organized attacks, and the impact of these crimes on financial stability, reputation, and critical infrastructure. The text emphasizes the importance of effective cybercrime detection in protecting sensitive data, preventing financial losses, and maintaining regulatory compliance. It provides a comprehensive overview of various types of cybercrimes, including malware attacks, phishing, identity theft, and financial fraud, and discusses the current trends and challenges in cybercrime detection. The chapter also outlines the objectives of cybercrime detection, such as educational purposes, knowledge dissemination, methodological guidance, technical insights, and best practices. It highlights the significance of feature selection in enhancing the accuracy of machine learning models used for cybercrime detection. The results section demonstrates the high accuracy and precision of these models in distinguishing between benign and malicious activities. The conclusion underscores the importance of high-quality datasets and meticulous data cleaning processes in building resilient cybercrime detection systems. This chapter offers valuable insights and practical recommendations for professionals seeking to enhance their cybersecurity strategies and stay ahead of emerging threats.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

13. SmartGuard: Advanced Security System with YOLOv8

    Vijayabhaskar Ch, V. Kakulapati, P. Devender Reddy, N. Manikanta Teja, J. Sri Vardan

    This chapter explores the integration of the SmartGuard security system with YOLOv8 for real-time object detection, enhancing surveillance capabilities. The text delves into the system's architecture, including its smart home features and automated door control, and evaluates its performance using metrics like accuracy, recall, and F1 score. It also discusses the implementation of an email alert system for proactive threat identification. The study concludes with potential future enhancements, such as incorporating infrared cameras and improving threat understanding. This comprehensive analysis provides insights into the system's efficiency and adaptability, making it a crucial read for professionals seeking to advance security measures.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

14. Comparative Study of Models in Sentiment Analysis

    Nagaratna P. Hegde, Sireesha Vikkurty, Sriperambuduri Vinay Kumar, Amruth Devineni, Sanjana Cherukuri

    This chapter presents a comparative study of various sentiment analysis models, including BERT, XLNet, Naive Bayes, CNN, and Logistic Regression, using a dataset of 50,000 movie reviews from IMDb. The study evaluates each model's performance using metrics such as F1 score, accuracy, precision, and recall. Notably, the ensemble method, which combines predictions from multiple models, achieves the highest recall, demonstrating its strength in capturing a broader range of sentiment instances. The chapter also discusses future works, including increasing epochs, scaling dataset size, integrating advanced models, optimizing ensemble approaches, and deploying models for real-world applications. The results highlight the diversity in performance across different models and underscore the potential of ensemble techniques in achieving comprehensive sentiment analysis outcomes.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

15. Intelligence Frameworks for Medical Image Analysis and Augmentation – A Review

    S. Radhika, G. Sharada

    This chapter delves into the cutting-edge realm of deep learning for medical image analysis, with a particular emphasis on lung disease diagnosis. It explores the evolution from traditional machine learning methods to advanced deep learning techniques, highlighting the role of convolutional neural networks (CNN) and recurrent neural networks (RNN) in enhancing diagnostic accuracy. The text reviews various state-of-the-art architectures such as Inception, ResNet, and DenseNet, and discusses their effectiveness in medical image processing. It also examines the use of hybrid learning models that combine CNN and RNN, which have shown promise in improving the classification and segmentation of medical images. The chapter further elaborates on data preprocessing, augmentation, and performance evaluation metrics, providing a comprehensive overview of the latest advancements in the field. Additionally, it discusses the practical applications of these techniques in healthcare, emphasizing their potential to revolutionize medical imaging and diagnosis.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

16. Field Oriented Control of PMSM for Flux Weakening Operation

    B. Naga Swetha, R. Geshma Kumari, K. Sravani, B. Harshini

    This chapter delves into the implementation of field-oriented control (FOC) for permanent magnet synchronous motors (PMSMs) in variable speed drives, focusing on flux weakening operation. The text begins by highlighting the advantages of variable speed drives, including reduced power line disruptions and regulated starting current, which have led to their widespread adoption in industries. The core of the chapter revolves around the design and modeling of a PMSM drive system using FOC, with a particular emphasis on reducing torque and speed ripples through sinusoidal pulse width modulation in MATLAB Simulink. The chapter also explores the use of flux weakening techniques to achieve higher speeds beyond the rated value, discussing the mathematical representation of the system and the role of discrete voltage vectors in minimizing flux and torque fluctuations. Through experimental results and simulations, the chapter demonstrates the dynamic performance of the PMSM drive system, showcasing its ability to maintain stable speed and torque under varying load conditions. Additionally, the chapter compares different control strategies, such as direct torque control (DTC), and highlights the benefits of FOC in achieving precise speed and torque control. The conclusion underscores the growing popularity of PMSM drives due to their efficiency and cost-effectiveness, positioning them as a superior alternative to conventional induction motors. This chapter provides a comprehensive overview of FOC techniques and their practical applications, making it an essential read for professionals seeking to enhance the performance of PMSM drive systems.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

17. Interactive Data Mining with Machine Learning: User-Centric Approaches and Tools

    Potlakayala Deepthi, Manchala Bhavani, Kasapaka RubenRaju, BommaReddy Sindhuja, Aluka Madhavi, Samala Nandini

    Interactive Data Mining with Machine Learning (IDM-ML) is transforming the way professionals analyze and interpret data. This chapter explores the synergy between human expertise and machine intelligence, emphasizing user-centric approaches that prioritize intuitive interfaces and graphical representations. Key topics include the integration of user interaction in the entire data mining process, from problem formulation to model interpretation, and the development of tools that cater to users with diverse backgrounds. The chapter also delves into the importance of interactive data exploration and visualization, feature selection and engineering, and model evaluation and validation. Additionally, it discusses the role of Human-in-the-Loop Machine Learning (HITL ML) in enhancing model interpretability and adaptability. The conclusion highlights the transformative potential of IDM-ML in reshaping the landscape of data mining and machine learning, making it more accessible, transparent, and user-friendly. Professionals will gain insights into practical tools and methodologies that can be applied across various domains, ultimately advancing the goal of creating effective machine learning solutions for complex decision-making scenarios.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

18. Smartmeals: A Dual Approach of Localized Dietary Recommendations and Predictive Model for Combating Child Malnutrition

    Tamminina Ammannamma, Vyjayanti Nandula, Ridhima Thakur, Saranya Gummireddy, Akshaya Juluri

    This chapter explores the development and implementation of an online platform designed to address child malnutrition in rural India. The platform leverages machine learning to provide personalized meal plans based on user-specific data, including age, height, weight, and location. A key feature is the BMI predictor, which forecasts BMI changes over 30 days, helping users maintain a healthy weight. The platform also offers location-specific recipe recommendations using locally available and affordable ingredients, ensuring accessibility and cultural relevance. Additionally, it supports multiple languages and audio input, making it user-friendly for individuals with varying levels of education. The chapter concludes with a real-world example from Andhra Pradesh, demonstrating the platform's effectiveness in suggesting appropriate recipes and predicting BMI with an accuracy of 80.5%. This innovative approach not only addresses the root causes of malnutrition but also provides a sustainable and scalable solution for improving child health outcomes in resource-constrained settings.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text