Babylonian Journal of Artificial Intelligence

A Survey on the Potential of Artificial Intelligence Tools in Tourism Information Services

Osamah Mohammed Alyasiri — 2024-01-10

Recently, ChatGPT, an advanced AI language model, has transformed the tourism industry by redefining how travelers access information and experience destinations. This survey explores the multifaceted potential of ChatGPT in various sectors of the tourism domain. The study begins by addressing the ethical considerations related to AI and natural language processing, emphasizing the necessity of privacy and data security in ChatGPT interactions. It analyzes the recent research findings, demonstrating ChatGPT’s efficacy in translating languages for tourists and improving services. The technology’s impact on tourism education and research is explored, highlighting its disruptive effects, benefits, and challenges, particularly in academic contexts. The paper delves into ChatGPT’s influence on the hospitality and tourism sector, focusing on its role in customer interactions, backend operations, and research methodologies. Additionally, it examines ChatGPT’s implications for content creation, visitor motivations, cultural perceptions, and regional tourism management. These insights shed light on ChatGPT’s potential to enhance customer experiences, influence visitor behaviors, and bridge cultural gaps in diverse tourism contexts. Concluding with preliminary guidelines for ChatGPT adoption, this paper equips industry professionals with essential knowledge to leverage the technology effectively. By embracing ChatGPT, the tourism industry can provide travelers with more informed, personalized, and immersive experiences, thus enhancing overall satisfaction and engagement during their journeys.

Improving Bus Passenger Flow Prediction Using Bi-LSTM Fusion Model and SMO Algorithm

Karthika Balasubramani — 2024-07-05

Bus passenger flow prediction integrates data analytics and modelling techniques to forecast the number of passengers using bus services, incorporating historical usage patterns, demographics, weather, and events for optimal scheduling and resource allocation. The Bi-LSTM fusion model enhances accuracy by processing past and future features simultaneously, leveraging bidirectional LSTM layers and an attention mechanism to capture temporal dependencies. This approach not only refines insights crucial for urban mobility challenges like traffic management and demand forecasting but also improves route planning and service efficiency. The SMO algorithm initializes with a diverse spider monkey population exploring solution spaces. Through local and global leader phases, it iteratively updates positions based on fitness and probabilistic selections, maintaining a balance between exploration and exploitation. Perturbation-based updates in the local leader phase ensure adaptability, preventing premature convergence, while the global leader phase guides towards better solutions, enhancing efficiency in complex optimization tasks and promoting dynamic adaptation. In Dataset 1, the proposed model achieved a training time of 137 seconds, slightly longer than HA (115s), SARIMA (112s), GRU (123s), and ST-ResNet (113s). It demonstrated superior accuracy at 89%, surpassing HA (66%), SARIMA (68%), GRU (63%), DeepST (78%), and ST-ResNet (84%). In Dataset 2, the model exhibited the lowest RMSE, MAE, and MAPE%, indicating superior predictive accuracy over SVR, CNN, GCN, LSTM, and CONV LSTM models. These findings validate the proposed model's effectiveness in enhancing predictive capabilities for transit forecasting, underscoring its potential to optimize urban mobility and transportation management strategies significantly.

Elderly People Health Care Monitoring System Using Internet of Things (IOT) For Exploratory Data Analysis

Rahul Sanmugam Gopi — 2024-06-02

The majority of senior persons are suffering a loss of physical condition their health is vital, the prevalence of illnesses has increased in the current age compared to previous generations, making health management in hospital sector. One of the concepts that can harness the advantages of the Internet of Things to better the older lifestyle is health monitoring for active and aided functioning. This approach aims to create a wireless communication system designed to remotely monitor patients. Each patient is wirelessly monitored based on Body temperature is measured via a temperature sensor, which gives an accuracy of around some certain value. The sensor carries a heart rate monitor with a pulse oximetry sensor. There are two light-emitting in it; one produces red light and the other infrared light. For tracking the amount of oxygen in the blood, two infrared LEDs are needed, but only one is needed to calculate the heart rate. The ESP 8266 (Node MCU) controller gathers the data and sends it to the specific application and Communication will be sent again to the consumers based on the data examination. The outcome results manage each patient's data separately and are used in hospitals, old age homes, and physically challenged people. Every sensor of data, such as temperature, heart rate, and abnormal heartbeat, will be managed in the database processed with GPS coordinates and data analyses using Think Speak Web Application is for informing about personal health data.

Generative Models in Natural Language Processing: A Comparative Study of ChatGPT and Gemini

Abdulazeez Alsajri — 2024-11-15

Generative models in natural language processing (NLP) have transformed machine interactions with human language. This research offers a comparative examination of ChatGPT and Gemini, two cutting-edge NLP models. This research elucidates the distinct strengths and shortcomings of each model by analyzing conversational fluency, diversity, and contextual accuracy. ChatGPT has exceptional originality in word production, but Gemini excels in contextual variety and sophisticated learning methodologies. The results highlight the revolutionary capability of generative models in fields such as education, customer service, and content production. Anticipated future improvements in this field are projected to enhance these capabilities, promoting wider societal and industrial advantages.

Enhanced Priority-Integrated Mult winner Voting Software

Firas Ali Hashim — 2024-04-07

Modern democracy values efficient, fair elections. This paper offers Google Cola an embedded interactive multi-winner voting method to improve democracy. Candidates and values can be entered using Python in the Colab. Basic Python algorithms and classes simplify priority candidate multi-winner elections. According to user feedback, event-driven programming modifies candidate priority enabling accurate and effective elections.

These embedded system functions prioritize, rank, and aggregate votes.

These mathematics facilitate fair and transparent outcome processing. This research demonstrates the embedded system's technology deployment and potential to support inclusive elections. The proposed solution helps electoral democratization by integrating collaboration with cutting-edge technologies

Authentication System Based on Fingerprint Using a New Technique for ROI selection

Wisam K. Jummar — 2024-08-25

Through the analysis of biological, behavioral, or a combination of both traits, biometrics entails the identification of specific persons. Finger veins, iris patterns, fingerprints, and DNA are examples of common biometric qualities. Of these, fingerprints are the most commonly utilized since they are highly distinctive, simple to obtain, and can be obtained from a variety of sources (because each person has ten fingers). In addition to introducing a method for identifying the region of interest (ROI), which is a specific area selected from the fingerprint image to improve feature extraction, this research focuses on several models for extracting fingerprint features. The study suggests a novel technique known as the flexible region of interest (FROI) for extracting a wider region of interest. Using the GLCM algorithm and invariant moments, features were retrieved from four fingerprint models (gray, binary, ridges thinning, and valleys thinning) using this FROI.

The highest performance was obtained with invariant moments recovered from the valleys thinning model, according to experimental results from a verification system. This led to a false rejection rate (FRR) of 7.5% and a false acceptance rate (FAR) of 0.3846%

Analyzing Ghana's Pharmacy Act, 1994 (Act 489) Regarding Quality Control and Negligence Liability Measures for Artificial Intelligence Pharmacy Systems

George Benneh Mensah — 2024-02-15

The objective of this systematic review was to assess the adequacy of current medication management in Ghana considering the risks posed by increased artificial intelligence (AI) automation in pharmacies worldwide A qualitative comparative approach was used despite reviewed the Ghana 1994 Pharmacy Act against recognition of AI challenges and international governance guidelines . The results revealed flaws in terms of quality prerequisites, transparency checklists and liability mechanisms developed for AI systems compared to existing regulations of the manual process. Outdated approaches to patient care that fail to ensure patient safety or address threats to the accuracy of recommendations from data collection biases and technical errors. Proposed changes include a requirement for usability testing before approving AI pharmacy deployments and the creation of a review board to review post-implementation systems for validity. Updating regulations to deal with modern equipment puts innovation and responsible regulation in the fast-paced healthcare industry. This study contributes significantly to preliminary research on AI policy readiness in the Ghanaian legal context, and suggests a feasible methodology for exploring qualitative differences for use in companies and countries competing for technology a disturbing, increasingly beyond the date code. Early government reform helps keep pace with the realities of adoption.

Predictive analytics model for students' grade prediction using machine learning

Muhammed Fareed Flayyih — 2024-08-05

This project aims to develop a predictive analytics model using machine learning to forecast student grades, helping educational institutions identify struggling students early for targeted support. By leveraging machine learning, the model can analyze large datasets to detect complex patterns, enhancing prediction accuracy in education. The project employs neural networks due to their ability to capture non-linear relationships in data. Two models were created: one trained with data from low-rated schools and tested on both low- and high-rated schools, achieving 85.7% and 83.3% accuracy, respectively. The second model, trained with high-rated school data, yielded 88.9% accuracy for high-rated schools but only 35.7% for low-rated ones. Results indicate that separate models for different school levels are more effective due to discrepancies in grade reporting accuracy among Iraqi schools.

A Review of Using Chatgpt for Scientific Manuscript Writing

Ahmed Adil Nafea — 2024-01-10

Scientific manuscripts an important in publishing research findings and advancing scientific knowledge. The method of writing a manuscript can take time and be a challenge for a lot of researchers. Artificial intelligence have helped the improvement of scientific writing. ChatGPT is an AI language model developed by OpenAI. This study review aims to evaluate ChatGPT for scientific manuscript writing, showing an overview of ChatGPT's abilities and limitations and highlighting its benefits and challenges in scientific writing. This study talks about review studied various methods in which researchers could use ChatGPT. This review also discussed ethical considerations related to ChatGPT in scientific manuscript writing including the impact of authorship, plagiarism, and the science community. It considers the importance of clarity and ensuring appropriate attribution when incorporating AI-generated content into scientific manuscripts the review gives the evaluation of researchers who using ChatGPT in scientific writing and it highlights their experiences challenges and recommendations for using ChatGPT as a written tool effectively.

Healthcare Analysis Based on Diabetes Prediction Using a Cuckoo-Based Deep Convolutional Long-Term Memory Algorithm

T. Kavitha — 2024-06-11

In recent years, the demand for mobile medical applications utilizing the Internet of Things (IoT) for diabetes diagnosis has been progressively increasing. Diabetes is commonly known as a chronic illness that presents a significant danger to individuals, analysing when blood sugar levels surpass the typical range. If diabetes is not promptly treated, hyperosmolar conditions can lead to serious health issues like hyperglycaemia and possibly even death. Since early detection enables lifestyle changes that prevent the disease's progression, it is crucial for diabetes management and health systems. However, diabetes diagnosis has a long computational time and low prediction accuracy. To address this issue, we propose a Cuckoo-Based Deep Convolutional Long-Term Memory (CDC-LSTM) algorithm that increases accuracy by classifying diabetics or non-diabetics. Additionally, we utilize the Standardized Feature Scaler (SFS) method to normalize the variance data by removing the mean of each feature. Moreover, we select the optimal features in the diabetes dataset utilising the Filter-Based Decision Tree (FBDT) technique. Finally, the proposed CDC-LSTM method can be used to distinguish between diabetics and non-diabetics, improving the accuracy of identifying diabetic patients. Additionally, the proposed method can predict diabetes using performance assessments such as precision, recall, and F-measure. Furthermore, the method's accuracy can be improved to 95.18% compared to previous approaches.

A Brief Review on Preprocessing Text in Arabic Language Dataset: Techniques and Challenges

Ahmed Adil Nafea — 2024-05-18

Text preprocessing plays an important role in natural language processing (NLP) tasks containing text classification, sentiment analysis, and machine translation. The preprocessing of Arabic text still presents unique challenges due to the language's rich morphology, complex grammar, and various character sets. This brief review studied various techniques utilized for preprocessing Arabic text data. This study discusses the challenges specific to Arabic text and current an overview of key preprocessing steps including normalization, tokenization, stemming, stop-word removal, and noise reduction. This survey analyzes preprocessing techniques on NLP tasks and focus on current research trends and future directions in Arabic text preprocessing.

Smart Wearables Powered by AI Transforming Human Activity Recognition

Vincent Omollo Nyangaresi — 2024-10-17

Smart clothing has changed the ways that human behaviour is observed and analyzed, finding its uses in health and fitness, and assisting in daily living. Nevertheless, conventional techniques used in HAR are mostly based on feature extraction by designers and the use of fixed algorithms that cannot address the dynamic aspects of human activities. HAR can be advanced through devices supported by artificial intelligence, and this research seeks to investigate how wearable technologies can improve this field of study. Hence, using CNN and RNN deep learning architectures this study constructs a comprehensive model with the potential of detecting various human activities instantaneously and accurately. The framework includes the use of sensor fusion approaches to process data collected from accelerometers, gyroscopes and heart rate sensors, to fully capture physical movements. Specifically, to high performance and efficiency in the computations of the model, several preprocessing and feature extraction techniques are employed. Outcome analysis shows that the proposed AI-based framework recognizes a subject’s identity with more than 95% accuracy across the different datasets comping basic machine learning techniques. The versatility of the system demonstrated in this work regarding wearable platforms and activities suggests that the proposed solution could be useful for practical application in fitness, health care, and rehabilitation. The findings of this work point to critical opportunities that are driven by AI for future wearable technology; a more adept and integrated system that can preserve the health and efficiency of humans. The results pave the way for further advancements in using AI to support HAR with a focus on scalability, real-time applicability, and inclusiveness.

Enhancing Agriculture Crop Classification with Deep Learning

Yasmin Makki Mohialden — 2024-03-02

To classify rice crops, the paper applies deep learning to agricultural crop images to classify rice crops. The collection includes images of wheat, rice, sugarcane, jute, and maize.

We improved variety by horizontally flipping, rotating, and shifting rice image data sets. A CNN structure classifies rice and non-rice crops.

The model has 100% accuracy on training and testing datasets; however, the classification report shows label imbalance problems for precision, recall, and F-score.

Deep learning can help classify crops as well as make decisions in agriculture based on research.

The study recommends more studies and improvements to enhance model performance and address dataset concerns. The research advances agricultural technology and emphasizes machine learning for crop management and production.

Enhancement of The Performance of Machine Learning Algorithms to Rival Deep Learning Algorithms in Predicting Stock Prices.

Rusul Mansoor Al-Amri — 2024-09-25

This paper objectives to improve stock market prediction accuracy by training data on sentiment analysis of tweets, overcoming volatility and complexity. Utilizing the Use of natural language processing (NLP) algorithms, the tweet's sentiments were classified into (negative - neutral - and positive). The stock value price was predicted by implementing Machine learning algorithms (KNN, SVM, GBM, LR, DT, RF, EL). Among the techniques of ML, (GBM) achieved the greatest results in terms of accuracy (96%). Its results were compared with the results of a deep learning algorithm that uses the same data where GBM got better results, and other algorithms showed results (KNN = 55%, SVM=90%, LR=82%, DT=90%, RF=90%, EL = 88%). The results obtained were superior to previous studies.