International Journal of Advances in Applied Computational Intelligence
Journal DOI
2833-5600ISSN (Online)
The cornerstone of crucial biochemical processes is enzymes, and this requires a need for precise detection methods to understand it all together and be able to intervene. This paper provides an innovative framework that addresses the problem of multi-label detection of enzyme-substrate interactions based on multi-label fusion. To overcome the limitations of traditional single-label detection approaches, our methodology combines several different data types and gradient boosting classifiers with CatBoost and AdaBoost classifiers as an ensemble. Our aim is to overcome the limitations of traditional single-label detection methods by integrating several data modalities and using a combination of Gradient Boosting, AdaBoost, and CatBoost classifiers. By means of comprehensive molecular descriptor analysis, clustering results, and model performance metrics visualization we demonstrate the intricate landscape of enzyme-substrate interactions in our research. Visualization techniques provide insights into the important molecular characteristics that influence the classes of enzymes while cluster analysis reveals inherent groupings within the dataset. The approach also employs confusion matrices to illustrate how well the model has been classified which supports the success of this framework. This method pushes forward multi-label information fusion as well as grounds for untangling biochemical complexities promising transformative applications across various scientific fields.
Read MoreDoi: https://doi.org/10.54216/IJAACI.050101
Vol. 5 Issue. 1 PP. 08-14, (2024)
This study presents a comprehensive evaluation of natural gas automobiles, focusing on their performance, environmental impact, economic viability, and potential as an alternative fuel for transportation. Natural gas vehicles (NGVs) have gained attention as an alternative to conventional gasoline or diesel vehicles due to their lower emissions profile and potential for reducing greenhouse gas emissions. The assessment encompasses a comparative analysis of NGVs against traditional internal combustion engine vehicles, evaluating factors such as vehicle efficiency, fuel availability, infrastructure, emissions, and cost-effectiveness. Findings reveal that NGVs exhibit lower emissions of pollutants like nitrogen oxides and particulate matter than their gasoline or diesel counterparts. However, challenges persist regarding limited refueling infrastructure, reduced driving range, and upfront vehicle conversion or purchase costs. Economic evaluations highlight the potential cost savings associated with natural gas as a fuel, particularly in regions with favorable pricing and infrastructure. Despite these benefits, scalability and widespread adoption of NGVs face barriers related to infrastructure development, technological advancements, and market incentives. This evaluation provides insights into the opportunities and challenges of natural gas automobiles, emphasizing the need for a balanced approach encompassing technological innovation, infrastructure investment, and supportive policies to unlock their full potential as a viable alternative in the transportation sector. We used multi-criteria decision-making (MCDM) to deal with various criteria of natural gas automobiles. The Range of Value Technique (ROV) method ranks the alternatives. The ROV is integrated with the neutrosophic set to deal with uncertainty information. The neutrosophic set is extension of fuzzy set to overcome the vague and incomplete information. The sensitivity analysis is conducted to check the stability of the results.
Read MoreDoi: https://doi.org/10.54216/IJAACI.050102
Vol. 5 Issue. 1 PP. 15-28, (2024)
Machine Learning (ML) and Artificial Intelligence (AI) are being employed in the fight against COVID19 by supporting the analysis of medical images, like X-rays and CT scans, to find characteristic paradigms linked with the virus. AI methods can evaluate huge volumes of data, which includes imaging data and patient medical records, for enriching the speed and precision of COVID19 diagnosis. Also, the use of ML and AI in medical imaging can aid in detecting new variants of viruses and forecasting their spread. The integration of ML and AI in COVID19 healthcare has greater potential to enhance the efficiency and accuracy of diagnoses along with that informing public health decision-making. Thus, the study proposes a Farmland Fertility Optimization Algorithm with Deep Learning based Healthcare Decision Making (FFOADL-HDM) approach for the detection of COVID19. The presented FFOADL-HDM approach emphasises the identification and classification of COVID19 using a CT scan. To achieve this, the FFOADL-HDM method exploits a modified SqueezeNet model for the generation of feature vector. Also, the hyperparameters of the modified SqueezeNet model can be selected by the use of FFOA. At last, the COVID-19 detection procedure is executed by the use of Adamax optimizer with (CFNN). The stimulation analysis of the FFOADL-HDM algorithm is studied on the SARS-CoV-2 CT image dataset from the Kaggle repository. The results highlighted the improved detection rate of the FFOADL-HDM technique over recent state of art approaches
Read MoreDoi: https://doi.org/10.54216/IJAACI.050103
Vol. 5 Issue. 1 PP. 29-39, (2024)
Intelligent data processing and mining of histopathological images involve the application of advanced techniques and algorithms to analyze and extract meaningful information from digital pathology images. Osteosarcoma is a general malignant bone cancer generally established in teenagers and children. Manual diagnoses of osteosarcoma is a laborious task and needs skilled professionals. The mortality rate can be minimalized only if it is identified on time. Automatic detection systems and new technologies were utilized to classify and analyze medical images that, minimalize the dependency on specialists and result in fast processing. Recently, a lot of Computer-Aided Diagnosis (CAD) systems were proposed by research workers to diagnose and segment osteosarcoma from medical images. Deep learning (DL) algorithms are employed for the automated recognition and identification of osteosarcoma on histopathological images (HSI). The study proposes an Improved Tunicate Swarm Algorithm with Deep Learning for Osteosarcoma Detection and Classification (ITSA-DLODC) approach on pathological imageries. The proposed ITSA-DLODC method mainly enhances the recognition and classification of osteosarcoma on HSI. To attain this, the presented ITSA-DLODC method performs feature extraction using ShuffleNet convolutional neural network model. Besides, the ITSA-based hyperparameter optimizer is exploited to finetune the hyperparameters of the ShuffleNet model. Moreover, the salp swarm algorithm (SSA) with convolutional autoencoder (CAE) approach was utilized for the recognition and identification of osteosarcoma. A wide range of analyses can be applied to exemplify the higher performance of the ITSA-DLODC methodology. The simulation study demonstrated the development of the ITSA-DLODC methodology over other present models
Read MoreDoi: https://doi.org/10.54216/IJAACI.050104
Vol. 5 Issue. 1 PP. 40-55, (2024)
Innovations in business intelligence are crucial in the digital era to staying popular and competitive across the increasing business trends. Businesses have started scrutinizing the next level of data analytics and business intelligence solutions. Customer Churn Prediction (CCP), on the other hand, a crucial for making business decisions, which correctly recognizes the churn customers and acts appropriately for customer retention. Customer churn is an unavoidable consequence when the user is not satisfied with the company’s service for a longer period. Service unsubscription by the user does not emerge unexpectedly; instead, it comes from the customer as a vigorous act owing to its accumulation of long-term disappointment. Thus, there is a need for the service provider to find and address their challenges related to customer satisfaction and service for retaining irate customers. The possibilities to predict customer churn have dramatically increased with the advances in artificial intelligence (AI) and machine learning (ML) algorithms. Therefore, this study introduces an Optimal Long Short Term Memory Based Customer Churn Prediction for Business Intelligence (OLSTM-CCPBI) method. The proposed OLSTM-CCPBI method incorporates many innovative components, such as Min-Max scaling for normalization, LSTM networks for temporal sequence modelling, and Adam optimization for hyperparameter tuning. The OLSTM-CCPBI method effectively captures temporal dependency in sequential customer data by leveraging the dynamic nature of the LSTM network, which enables correct prediction of churn events. Through detailed investigations on real-time customer churn datasets, OLSTM-CCPBI achieves better predictive capabilities than classical approaches, showcasing its promising solution to aid businesses in preemptively addressing customer attrition and considerably enhancing churn prediction accuracy.
Read MoreDoi: https://doi.org/10.54216/IJAACI.050105
Vol. 5 Issue. 1 PP. 56-64, (2024)