Heart failure is a leading cause of death among Diabetic and Obese patients globally, contributing to 8.5% of all heart disease deaths and potentially 36% of cardiovascular disease deaths. Early detection is crucial for timely intervention, reducing symptoms, lowering hospitalizations, and improving patient outcomes through personalized management. This paper presents the use of Semi-Parametric and Accelerated Failure Time survival models (AFT) on Heart failure prediction. Cox Proportional Hazard model from the family of Semi-Parametric survival models and Weibull's Accelerated Failure Time models from the family of Accelerated Failure Time models have been compared and contrasted. The Cox PH model excels in its ability to adapt to various survival time distributions avoiding any assumptions on survival time distribution. The Cox proportional hazard allows for the examination of covariate effects on the hazard function, making it a widely used survival model. Weibull AFT model follows a parametric approach, directly estimating the distribution of the survival time. Medical records data of close to 299 patients, who had heart failure, collected during their clinical followup period is used for building and evaluating the model. The final evaluation of model performance was conducted, focusing on their capacity to predict the probability of patient survival beyond 250 days from the clinical visit. Among the Cox proportional hazard model and Weibull's AFT model, the Weibull's AFT model demonstrated superior performance compared to the Cox model. Remarkably, Weibull's model exhibited consistently exceptional performance in both train and test validations.

Understanding a product's image within a specific market is essential for evaluating and developing effective internal and competitive strategies to enhance product performance. This article discusses the application of Machine Learning (ML) techniques for understanding and proactively predicting a drug's performance rating across multiple dimensions. The ML models developed in this research can analyze shifting patient sentiments toward drugs over time, enabling pharmaceutical organizations to take proactive measures in response to changes in patient sentiment within a given market. Multiple ML algorithms were explored for this purpose, including Support Vector Machine (SVM), Deep Neural Networks, and Bidirectional Encoder Representations from Transformers (BERT). The performance of each algorithm was evaluated on both training and testing datasets sourced from the UCI Machine Learning Repository. The algorithm demonstrating the best performance has been selected for implementation as a front-end application for real-time drug rating prediction based on review text.

Quality assurance in manufacturing often involves manual interventions; however, the integration of machine learning and artificial intelligence can facilitate automation in quality inspection. Industrial process automation significantly contributes to the success of large-scale industries by improving operational efficiency and reallocating resources towards customer satisfaction, thereby enhancing overall business performance. This paper introduces a Convolutional Neural Network (CNN) model based on transfer learning for identifying surface irregularities on finished wooden objects, including stains, porosity, color mismatches, cracks, and knots. Unlike traditional methods that utilize intensity or colored images, our model employs image gradients based on the Canny operator. Edge detection is applied to raw colored images during the image augmentation process to accentuate structural changes on the objects. The rationale behind developing the model on edge-detected images is to reduce noise by eliminating natural patterns that could be misinterpreted as surface irregularities. Deep Convolutional Neural Network models are trained on these edge-detected images. The developed model is assessed using an out-ofsample dataset, with the F1-Score and Accuracy as the primary evaluation metrics. Initially, the model faced challenges in out-of-sample validation due to a low volume of defective images. However, post-performing class balancing, the model exhibited exceptional performance, highlighting its effectiveness in wooden surface irregularity identification.

Product pricing is a critical task that has a profound impact on demand and the target audience. Setting the price of an existing product is even more challenging, as it can significantly affect business growth, customer purchasing patterns, and brand perception. In this paper, we propose an approach that leverages a deep learning Multi-layer Perceptron (MLP) Neural Network to predict changes in drug prices. Our model differs from existing approaches as it focuses on utilizing the stated reasons for price changes rather than product or market attributes for price change prediction. The MLP-based model is designed to learn complex patterns and dependencies within the drug price data, providing a data-driven solution for accurate forecasting. The text field containing the reason for the price change is pre-processed and then fed into the MLP neural network. We employ a Continuous Bag of Words approach to convert the text into a numerical format for model development. The model is trained on a diverse dataset, and its performance is evaluated using various metrics, including Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), which indicate the model’s predictive accuracy. The results demonstrate that the MLP-based model excels in predicting fluctuations in drug prices, showcasing its ability to adapt to the dynamic nature of the pharmaceutical market.

Real estate prices are an important reflection of the economy and their prices are great interest to both buyers and sellers. Hundreds of houses are sold every day and the buyer asks himself what is the reasonable price that this house deserves. In this paper, a new regression model is proposed for the accurate prediction of house prices. This model is based on the hybrid recurrent neural network where the Gated Recurrent Unit (GRU) is fused with the Long Short-Term Memory (LSTM) and applied to a particular dataset that characterizes houses in Boston. Massachusetts dataset from Scikit-learn is used in this research to train and evaluate this regression model using the data on Boston housing. Several experiments were conducted on the proposed algorithm and evaluated with the commonly used metrics. The results of these experiments showed that the proposed model has better performance when the networks are used in the fusion process than when they act individually. It also has better accuracy and lower Root Mean Square Error when compared to several states of art methodologies.