International Journal of Advances in Applied Computational Intelligence
Journal DOI
2833-5600ISSN (Online)
Stock is a financial product considered by flexible trading, high risk, and high return that can preferred by several investors. Investors may get an abundance of returns through the accurate prediction of stock price trends. Nevertheless, the stock price can be influenced by certain factors including market conditions, companies’ managerial decisions, macroeconomic situation, and investors’ preferences for major economic and social events. Econometric and Statistical models are widely utilized in classical stock price prediction; however, these techniques could not handle the complex and dynamic environments of the stock market. Researchers have begun using deep learning (DL) and machine learning (ML) to estimate stock fluctuations and prices with the rapid evolution of artificial intelligence (AI), serving investors to define investment strategies to increase returns and decrease risk. Therefore, this manuscript presents a new dung beetle optimization with deep learning based stock price prediction (DBODL-SPP) methodology. The purpose of the DBODL-SPP algorithm is to predict the rise or fall of stock prices using the optimal DL model. In the DBODL-SPP technique, the min-max scalar can be deployed for pre-processing the input data. Besides, the DBODL-SPP approach applies the DBO algorithm for electing an optimal subset of features. The DBODL-SPP technique makes use of a multi-head attention long short-term memory (MHA-LSTM) model for the stock price prediction. Finally, by using the equilibrium optimizer (EO) algorithm, the parameter tuning of the MHA-LSTM algorithm can be carried out. A detailed set of experimentations has been applied to evaluate the enriched performance of the DBODL-SPP technique. The simulation values emphasized that the DBODL-SPP algorithm achieves better results than other techniques for stock price prediction
Read MoreDoi: https://doi.org/10.54216/IJAACI.050201
Vol. 5 Issue. 2 PP. 08-23, (2024)
Social isolation and loneliness are subjective measures related to the feeling of distress and discomfort for disabled and elderly people. Currently, computing platform offers a smart healthcare observing technique for earlier fall detection. Internet of Things (IoT) based health system had a crucial role in the healthcare service and assists in improving data processing and its prediction. Transmitting data or reports takes more energy and time, as well as causes energy issues and higher latency. These study concentrations on the development of Whale Optimization Algorithm with Deep Learning based Indoor Monitoring System (WOADL-IMS) for Elderly and Disabled People. The presented WOADL-IMS system purposes to identify the presence of indoor activity by elderly people. In the presented WOADL-IMS technique, NASNetMobile model is applied to produce feature vectors. In addition, the WOADL-IMS technique uses WOA based hyperparameter selection approach. Finally, triplet neural network (TNN) model can be employed for automated classification and recognition of indoor activity. The simulation result of the WOADL-IMS approach can be examined on indoor activity dataset. The outcomes of the experimentation highlighted that the WOADL-IMS technique reaches better results than other recent approaches
Read MoreDoi: https://doi.org/10.54216/IJAACI.050202
Vol. 5 Issue. 2 PP. 24-33, (2024)
With the large-scale language methods namely ChatGPT, there is a chance to explore the use of machine learning (ML) methods on ChatGPT-generated data for classifying healthcare data. Healthcare data classification gains more significance in extracting and organizing useful insights from the huge volume of medical data available. The ChatGPT-generated data has realistic and different healthcare-based text datasets that can be applied to training classification methods. ML approaches include supervised learning methods as support vector machines (SVMs), and random forests (RF), which can be implemented for classifying the healthcare data. The methods were trained on the ChatGPT-generated data that can be carefully validated and labelled with suitable classes related to the healthcare field. With this motivation, this article presents an automated healthcare data classification-using barnacles mating optimizer with a pyramid neural network (AHDC-BMOPNN) technique. The presented AHDC-BMOPNN technique examines the healthcare data effectually using an ML model with a feature selection process. Primarily, the AHDC-BMOPNN technique exploits min-max data normalization for scaling the input dataset. In addition, the butterfly optimization algorithm-based feature selection (BOA-FS) method is deployed for the selection of optimum feature subset. In this work, the PNN algorithm was utilized for the classification of medical data. Ultimately, the BMO-based hyperparameter tuning process takes place to boost the overall classifier results of the PNN technique. The empirical findings of the AHDC-BMOPNN approach was validated on ChatGPT generated dataset. The simulation values highlight that the AHDC-BMOPNN method and the diverse healthcare text data generated by ChatGPT enhance the ability to extract valuable insights and organize medical information effectively.
Read MoreDoi: https://doi.org/10.54216/IJAACI.050203
Vol. 5 Issue. 2 PP. 34-45, (2024)
Deception detection means finding whether an individual is lying or being deceptive depending on cognitive cues, and various behavioural, or physiological. It is a significant domain of research with applications in social psychology, law enforcement, and security. Deception detection relevant to microexpressions includes examining these subtle facial cues for determining whether an individual is being deceptive or lying. Microexpressions can deliver significant cues to detect deception. Deep learning (DL) and Machine learning (ML) models were utilized for finding micro-expressions and are trained for differentiating deceptive statements from genuine ones. Still, it necessitates a diverse and large dataset of video recordings in addition to careful tuning and pre-processing of the DL approach. So, this article presents an Automated Deception Detection on Facial Microexpressions using Improved Sparrow Swarm Optimization with Deep Learning (ADDFM-ISSODL) method. The proposed ADDFM-ISSODL algorithm examines facial micro-expressions effectively for detection of deceptive behaviour. To complete this, developed ADDFM-ISSODL model uses a Gaussian filtering (GF) approach for pre-processing. Besides, ADDFM-ISSODL technique employs MobileNetv3 model for feature extraction and the hyper parameter tuning procedure performed using ISSO algorithm. The ISSO approach was designed by the integration of the standard SSO approach with the quantum evolutionary algorithm (QEA). For deception detection, a probabilistic neural network (PNN) classifier was employed. At last, grasshopper optimization algorithm (GOA) was implemented for parameter tuning of PNN method. The performance validation of ADDFM-ISSODL system tested utilizing facial expression dataset. The simulation outcome stated the greater results of ADDFM-ISSODL algorithm over other methodologies.
Read MoreDoi: https://doi.org/10.54216/IJAACI.050204
Vol. 5 Issue. 2 PP. 46-59, (2024)
Multi-criteria decision-making (MCDM) is employed for analyzing traffic in a Vehicle-to-Everything (V2X) network. V2X suggests communication among vehicles and other entities, containing pedestrians, infrastructure, and other vehicles. Traffic analysis and management in V2X networks need effectual decision-making approaches, which assume several conditions. MCDM contains estimating and choosing alternatives depending on several conditions or objectives. In the context of traffic analysis in V2X networks, MCDM algorithms are employed for decision-making concerning traffic flow optimizer, resource allocation, route planning, and congestion management. Deep learning (DL) approaches are trained to analyze massive counts of data gathered from several sources from the V2X network. These sources contain traffic sensors, GPS data, vehicle-to-infrastructure (V2I) communication, and historical traffic designs. By processing this data, DL approaches extract useful insights and create informed decisions depending on various conditions. Therefore, this article proposes a gorilla troops optimizer with deep learning-based MCDM for traffic analysis (GTODL-MCDMTA) technique in the V2X network. The purpose of the GTODL-MCDMTA algorithm is to identify the traffic flow prediction for improving route planning and resource allocation with the consideration of various factors into account. In the presented GTODL-MCDMTA technique, the input data is pre-processed to remove noise and normalize it for analysis. Next, the GTO algorithm is used for the feature selection process. Besides, the deep extreme learning machine (DELM) model is used for the forecast of traffic movement. Finally, the seeker optimization algorithm (SOA) has been utilized for the parameter tuning of the DELM technique. A brief set of simulation outcomes can be applied to emphasize the promising outcomes of the GTODL-MCDMTA technique. The experimental outcome demonstrates the efficiency and efficiency of the GTODL-MCDMTA approach in handling the complexity and dynamic nature of V2X network traffic analysis.
Read MoreDoi: https://doi.org/10.54216/IJAACI.050205
Vol. 5 Issue. 2 PP. 60-72, (2024)