October 07 ~ 08, 2023, Virtual Conference
Adekunle Adefabi, Somtobe Olisah, Callistus Obunadike, Oluwatosin Oyetubo, Esther Taiwo, Edward Tella, Department of Computer Science and Quantitative Methods, Austin Peay State University, Tennessee
Road accidents have significant economic and societal costs, with a small number of severe accidents accounting for a large portion of these costs. Predicting accident severity can help in the proactive approach to road safety by identifying potential unsafe road conditions and taking well-informed actions to reduce the number of severe accidents. This study investigates the effectiveness of the Random Forest machine learning algorithm for predicting the severity of an accident. The model is trained on a dataset of accident records from a large metropolitan area and evaluated using various metrics. Hyperparameters and feature selection are optimized to improve the model's performance. The results show that the Random Forest model is an effective tool for predicting accident severity with an accuracy of over 80%. The study also identifies the top six most important variables in the model, which include wind speed, pressure, humidity, visibility, clear conditions, and cloud cover. The fitted model has an AUC of 80%, a recall of 79.2%, a precision of 97.1%, and an F1 score of 87.3%. These results suggest that the proposed model has higher performance in explaining the target variable, which is the accident severity class. Overall, the study provides evidence that the Random Forest model is a viable and reliable tool for predicting accident severity and can be used to help reduce the number of fatalities and injuries due to road accidents in the United States.
Machine Learning, Random Forest Model, Accident Severity Prediction, Mean Decrease Gini.
M.Gokilavani, Sriram and S.P.Vijayaragavan, Dept of CSE, BIHER, Chennai, Tamilnadu, India
Cancerous is an almost deadly disease that con- sequences, when cellular growth result in uncontrolled tumour and uncertainty of human cells. Around are numerous of types like skin, breast, bone, brain, etc. cause evident growing called benign tumour, although others, like blood, thyroid, lung, do not. Malignant cells may appear in unique region, formerly spread via lymph nodes. These are collections of immune cells placed all over the body. In medical field, cancer diseases is the well challenging disease in diagnosing for the removal and effective treatment methods. ML techniques like Deep learning, neural networks (NN) and fuzzy logic (FL) algorithms that have massive applications in the computerization of a process. Deeping learning plays the very effective role in early diagnosis to improve greater accuracy using various machine learning and deep learning architecture. Very promising machine learning algorithms with high-performance computing provides capable outcomes in medical imaging analysis like image fusion, image segmentation, and image registration and image classification. In this survey we focused on major role of ML and deep learning algorithm in thyroid and lung cancer diagnosis specifically in the CT/MRI scan images and also in the detection of lymph nodes characteristics leads to metastatic. The summary about Deep learning in thyroid and lung cancer diagnosis, challenges and future scope is also emphasized in this paper.
Deep learning and ML algorithms, Thyroid and Lung cancer.
Anusuya Krishnanl1 and Kennedyraj2, 1College of IT, UAE University, Al Ain, UAE, 2College of IT, Noorul Islam University, Kanyakumari, India
Cancerous is an almost deadly disease that con- sequences, when cellular growth result in uncontrolled tumour and uncertainty of human cells. Around are numerous of types like skin, breast, bone, brain, etc. cause evident growing called benign tumour, although others, like blood, thyroid, lung, do not. Malignant cells may appear in unique region, formerly spread via lymph nodes. These are collections of immune cells placed all over the body. In medical field, cancer diseases is the well challenging disease in diagnosing for the removal and effective treatment methods. ML techniques like Deep learning, neural networks (NN) and fuzzy logic (FL) algorithms that have massive applications in the computerization of a process. Deeping learning plays the very effective role in early diagnosis to improve greater accuracy using various machine learning and deep learning architecture. Very promising machine learning algorithms with high-performance computing provides capable outcomes in medical imaging analysis like image fusion, image segmentation, and image registration and image classification. In this survey we focused on major role of ML and deep learning algorithm in thyroid and lung cancer diagnosis specifically in the CT/MRI scan images and also in the detection of lymph nodes characteristics leads to metastatic. The summary about Deep learning in thyroid and lung cancer diagnosis, challenges and future scope is also emphasized in this paper.The exponential growth of online social network platforms and applications has led to a staggering volume of user-generated textual content, including comments and reviews. Consequently, users often face difficulties in extracting valuable insights or relevant information from such content. To address this challenge, machine learning and natural language processing algorithms have been deployed to analyze the vast amount of textual data available online. In recent years, topic modeling techniques have gained significant popularity in this domain. In this study, we comprehensively examine and compare five frequently used topic modeling methods specifically applied to customer reviews. The methods under investigation are latent semantic analysis (LSA), latent Dirichlet allocation (LDA), non-negative matrix factorization (NMF), pachinko allocation model (PAM), Top2Vec, and BERTopic. By practically demonstrating their benefits in detecting important topics, we aim to highlight their efficacy in real-world scenarios. To evaluate the performance of these topic modeling methods, we carefully select two textual datasets. The evaluation is based on standard statistical evaluation metrics such as topic coherence score. Our findings reveal that BERTopic consistently yield more meaningful extracted topics and achieve favorable results.
Natural Language Processing, Topic Modeling & Customer Reviews.
Veronica Mendes Pinto1,2, Tiago Gonçalves Silva1,2,3 and José Silvestre Silva1,3,4, 1Portuguese Military Academy, Lisbon, Portugal, 2Guarda Nacional Republicana (GNR), Lisbon, Portugal, 3Military Academy Research Center (CINAMIL), Lisbon, Portugal, 4LIBPhys-UC & LA-Real, Universidade de Coimbra, Portugal
The objective of this work is to support the continuous development and modernization of the Gendarmerie Forces, analyzing the use of Drones in the police activities of the Portuguese and Spanish Gendarmerie forces, examining the extent to which the implementation and expansion of these resources represent an advantage for the police service and operations of the Gendarmerie. Due to the great changes taking place in the world, it is crucial to rethink state security. The convergence of internal and external threats, together with the increase in the feeling of insecurity on a global scale, are emerging factors that require different strategies of the security forces, especially with regard to the support tools used. Thus, there was a need to analyze the legal panorama of the Drones used in police, military and customs missions, verifying their classification as state aircraft, verifying the current doctrine (civil, military and police), taking into account the analysis of these scenarios. To this end, a methodology based on the inductive method was adopted, which made it possible to generalize the data collected through the analysis of data on the Spanish Gendarmerie force, appreciating their characteristics and use, with the aim of comparing the modus operandi with the Portuguese Gendarmerie force.
Drones; Unmanned Aircraft Systems; Unmanned Aerial Vehicle; Guarda Nacional Republicana; Guardia Civil.
Taiwo Esther, Akinsola Ahmed, Tella Edward, Makinde Kolade, Akinwande Mayowa, Department of Computer Science, Austin Peay State University, Clarksville USA.
Thisstudy is focused on the ethics of Artificial Intelligence and its application in the United States, the paper highlights the impact AI has in every sector of the US economy and multiple facets of the technological space and the resultant effect on entities spanning businesses, government, academia, and civil society. There is a need for ethical considerations as these entities are beginning to depend on the use of AI for delivering various crucial tasks which immensely influence their operations, decision-making and interactions amongst each other. The adoption of ethical principles, guidelines and standards of work is therefore required throughout the entire process of AI development, deployment, and usage to ensure responsible and ethical AI practices. Our discussion explores eleven fundamental 'ethical principles' structured as overarching themes. These encompass Transparency, Justice, Fairness, Equity, NonMaleficence, Responsibility, Accountability, Privacy, Beneficence, Freedom, Autonomy, Trust, Dignity, Sustainability, and Solidarity. These principles collectively serve as a guiding framework, directing the ethical path for the responsible development, deployment, and utilization of artificial intelligence (AI) technologies across diverse sectors and entities within the United States. The paper also discusses the revolutionary impact of AI applications, such as Machine Learning, and explores various approaches used to implement AI ethics. This examination is crucial to address the growing concerns surrounding the inherent risks associated with the widespread use of artificial intelligence.
Ethics, Artificial Intelligence, Machine Learning, Technology, Ethical Principles.
Unuriode Austine, Durojaiye Olalekan, Yusuf Babatunde, Okunade Lateef, Department of Computer Science, Austin Peay State University, Clarksville USA.
Artificial intelligence (AI) has rapidly transformed various domains, reshaping how we interact with technology and access information. The pervasive influence of AI technologies spans all aspects of contemporary life. Over the years, the development of AI has been a journey marked by significant milestones, with numerous AI applications finding their way into diverse systems and applications [1]. While AI's impact is evident in many areas, one particularly intriguing and evolving frontier is the fusion of artificial intelligence with database systems. This amalgamation represents a recent and experimental endeavor with the potential to revolutionize database management's future. Recent progress in this field has laid the foundation for more intelligent and efficient database systems. In this comprehensive review, we embark on an exploration of the dynamic landscape of AI and database integration, delving into key areas of advancement and the research initiatives that have contributed to shaping this exciting field, Some of the areas are: (a) The design of an Intelligent Database Interface (IDI) which will be a huge growth in the database [2,3, 4]. (b) Learnable databases; how machine learning will further foster the advancement of AI/DB integration [12]. (c) The smart Query [5]. The Intelligent Database Interface (IDI) is a crucial component of AI and database integration [6]. It acts as a vital link connecting users to various databases, offering an interactive and efficient experience. Key elements include the Intelligent Database Interface Language (IDIL), which translates queries into SQL for the appropriate DBMS [6]. Learnable Databases represent a transformative shift by integrating machine learning (ML) into database systems. They have the unique capability to learn from historical data, making databases more intelligent and adaptable [7]. Smart Queries are designed to enable natural language interactions with databases, catering to users without programming expertise. This subfield of Natural Language Processing (NLP) aims to simplify human-computer interactions within the database domain [4].
Artificial Intelligence (AI), Database, IDI, SQL, Machine learning, NLP, Cloud.
SuvarnaVani, Harshitha Badavathula, Prasanna Vadttitya, Sujana Sri Kosaraju, Department of Computer Science and Engineering, Velagapudi Ramakrishna Siddhartha Engineering College, Kanuru, Vijayawada, Andhra Pradesh, India.
In hilly places, landslides cause the most severe geological risk. The ability to predict landslides is essential for saving lives in mountainous areas. So, in order to do this, we must collect historical data. Literature cannot heal places where there is no knowledge of their history. In these circumstances, machine learning can solve the issue, but choosing the most appropriate approach is difficult due to the wide variety of machine learning algorithms. As a result, it's important to compare and choose the machine learning techniques that produce the best outcomes. The effectiveness of a machine learning method called a convolutional neural network is utilized in this study to forecast landslides. There will be a vast data set because the landslides are happening at various times and places around the world. Transfer learning can be used to organize and train the huge dataset. It carries out object detection, segments the images, and measures the colors and shapes of the images. The given input is compared using these images. As a result, the study is successful in comparing and illustrating machine learning algorithms for landslide prediction.
Machine Learning, Convolutional Neural Network, Satellite Images Transfer Learning, Webpage (Flask App).
A. Sai Kumar, B. Jagadeesh Sai, G. Rushivardhan Babu, A.M.L. Narayana, Pooja Panapana, Department of Information Technology, GMR Institute of Technology, Andhra Pradesh, India.
The rapid evolution of Human-Computer Interaction (HCI) has witnessed the amalgamation of gesture recognition into various applications, transforming the way humans engage with machines. This paper presents a comprehensive review focusing on the integration of gesture and body posture recognition in both gaming and surveillance sectors. We examine the innovative use of gesture recognition in rejuvenating classic games, such as the Snake game, wherein gestures replace traditional control keys, offering a more interactive experience. Drawing parallels, we delve into the realm of human-robot collaboration, exemplified by the Rock-Paper-Scissors game, highlighting the blend of computational efficiency and engaging human-like responses from robots. Beyond gaming, the paper underscores the significance of vision-based body posture detection in surveillance systems, discussing its vast applications, ranging from basic recognition to intricate behaviour analysis. Through a detailed exploration of methodologies, such as OpenCV's integration in gaming and the utilization of OpenPose MobileNet Technology in posture detection, we offer insights into the strengths and challenges each approach presents. In synthesizing these diverse applications, this survey underscores the transformative potential of gesture and posture recognition, illuminating pathways for future research and application in HCI.
OpenPose MobileNet Technology, Computer-Vision, Gesture and Posture Recognition, Human-Computer Interaction.