Journal of Data Mining and Management (e-ISSN: 2456-9437)

Security Challenges in Data Collection and Processing in Industry 4.0 Implementation

N. Vamsi Krishna — 2023-09-19

IoT is crucial to the implementation of Industry 4.0. Security is an important factor to consider while managing data. At the same time, the Internet of Things (IoT) is a rapidly evolving technological paradigm that promises to revolutionize the way people interact with the world around us. It involves the integration of various devices and sensors into everyday objects, enabling them to collect, exchange, and analyze data to enhance convenience and efficiency. The applications of IoT are vast and diverse, encompassing smartwatches, smartphones, industrial processes, and even educational settings. Central to the functioning of IoT is the seamless exchange of information among interconnected devices. However, this exchange often includes personal and sensitive data, making security a paramount concern. Protecting this data is essential to prevent potential security threats and breaches. This paper delves into the multifaceted world of IoT, exploring its applications across various domains while shedding light on the security challenges it presents. It delves into different types of security threats that can compromise the integrity and confidentiality of IoT data, such as unauthorized access, data breaches, and device manipulation. Moreover, the paper also provides insights into strategies and technologies to mitigate these risks. It discusses the importance of robust authentication protocols, encryption mechanisms, and intrusion detection systems to safeguard IoT ecosystems. As the IoT continues to grow and intertwine with our daily lives, addressing security concerns is crucial to fully harness its potential while ensuring the safety and privacy of individuals and organizations alike.

A Survey on Diseases of Grape Leaf Identification Using Techniques of Machine Learning

Noorjahan M. Attar — 2023-12-21

This paper presents a survey of different types of leaf diseases of the grapes and introduces variant kinds of techniques that are the detection of diseases. Grape diseases could cause financial losses to farmers if not detected. Grape leaves are affected by fungal infection, bacterial infection, and viral infection. Target diseases are black rot, black measles, white powdery, powdery mildew, downy mildew, and anthracnose. Detecting the diseases in leaves is the most precious method for increasing food production. It is very difficult and time-consuming to follow and test grape diseases manually. Effective detection of grape leaf diseases is crucial for preventing financial losses to farmers. This paper conducts a comprehensive survey of various grape diseases, including fungal, bacterial, and viral infections such as black rot, black measles, white powdery, powdery mildew, downy mildew, and anthracnose. To address the challenges posed by manual inspection, the paper introduces diverse techniques for disease detection. Automating this process not only enhances efficiency but also plays a pivotal role in boosting food production, emphasizing the significance of advanced methods in agriculture.

Recommendation System to Precision Agriculture Using Machine Learning Algorithm

Jannu Dhanusha — 2023-12-21

The Indian economy heavily relies on agriculture, a vital source of livelihood for millions of farmers and a significant contributor to the GDP. However, farmers often face challenges in selecting the most suitable crops for their specific soil and environmental conditions. This decision profoundly influences crop productivity and the overall agricultural industry. Addressing this issue is crucial, and crop prediction plays a pivotal role in the emerging concept of precision agriculture. Precision agriculture seeks to streamline the crop selection process by leveraging machine learning techniques. The model takes into account various factors such as soil type, rainfall patterns, temperature, groundwater levels, availability of pesticides and fertilizers, and the current season. These elements are intricately analyzed to create a robust recommender system. The ultimate objective is to expedite and enhance the crop selection procedure for farmers. By incorporating data-driven insights, this precision agriculture approach aims to empower farmers with informed decisions tailored to their unique agricultural landscapes. The recommender system acts as a valuable tool, offering guidance on the most suitable crops for optimal yield and sustainability. This initiative not only addresses the challenges faced by Indian farmers but also contributes to the overall efficiency and resilience of the agricultural sector. In summary, the integration of machine learning and data analytics in crop prediction for precision agriculture holds the potential to revolutionize decision-making processes and positively impact crop productivity in the Indian agricultural landscape.

A Qualitative Survey of Recent Advances in Artificial Intelligence

V. Deepa — 2023-12-26

More than 60 years ago, the phrase "artificial intelligence" was first used. Artificial intelligence (AI), machine learning, robotics, and automation are exponentially advancing fields that are fast-changing global economies and civilizations. In many areas, including education, weather forecasting, agriculture, and others, AI is now a part of daily life. The majority of us already employ intelligent devices that can learn, fix issues, and offer suggestions on everything from driving routes to clothing purchases. We have robots in our workplaces, self-driving cars on our highways, and intelligent personal assistants sitting on our countertops. And that is only the beginning. AI has advanced significantly during the last five years. The developments in AI are outlined in this qualitative survey article.

A Systematic Analysis in Social Media Platforms for Cyberbullying Detecting System Using Machine Learning Techniques

T. Sarathamani — 2023-12-30

Social media platforms offer us more opportunities, than ever before and it is undeniable that they bring numerous benefits. However, it is important to confess that societies can still face humiliation, disdain, intimidation and harm from individuals or even their earls. Cyberbullying stands for the usage of technologies to defame and belittle others often through messages sent via media or email. With the growth of social media users, cyberbullying has also arisen in the method of email bullying. We undertook a project aimed at addressing this issue by examining cyberbullying through tweets using ML reinforcement algorithms such as Naive Bayes, KNN; Decision Tree, Random Forests, and Support Vector Machines (SVM). Additionally, we applied the NLTK Natural Language Toolkit techniques such as bigram analysis, trigram analysis gram analysis unigram analysis and n Naive Bayes to assess its consistency. Finally, a thorough examination was conducted on the outcomes of implementing machine learning algorithms for detecting cyberbullying while considering features and propositions. The results of our analysis demonstrate the implication of future work in identifying instances of cyberbullying.