Triazolopyrimidine and its analogues represent important lead structures in anti-malarial research. This study modelled the quantitative structure–activity relationship (QSAR) of 125 congeners of 1,2,4-triazolo[1,5-a]pyrimidine from the ChEMBL database in the inhibition of Plasmodium falciparum using six machine learning algorithms. Recursive feature elimination was used to select the most significant of 306 molecular descriptors with 1 moleculal outlier. A split ratio of 20 % was used to split the x and y matrices into 99 training and 24 test compounds. The regression models were built using machine learning scikit-learn algorithms (multiple linear regression (MLR), k-nearest neighbours (kNN), support vector regressor (SVR), random forest regressor (RFR), RIDGE regression and LASSO). Model performance was evaluated using the coefficient of determination (R²⁾, mean squared error (MSE), mean absolute error (MAE) and root mean squared error (RMSE) p-values, F-statistic and variance inflation factor (VIF). The number of significant variables considered to build the model was 5 (p < 0.05) with the regression equation pIC₅₀ = 5.90–0.71npr1–1.52pmi3+0.88slogP–0.57vsurf-CW2+1.11vsurf-W2. On 5-fold cross validation, three algorithms, KNN (MSE=0.46, R²=0.54, MAE=0.54, RMSE=0.68), SVR (MSE=0.33, R²=0.67, MAE=0.46, RMSE=0.57) and RFR (MSE=0.43, R²=0.58, MAE=0.51, RMSE=0.66) showed robustness, high efficiency, and reliability in predicting the pIC₅₀ of 1,2,4-triazolo[1,5-a]pyrimidine. The models provided useful data on the functionalities necessary for developing more potent 1,2,4-triazolo[1,5-a]pyrimidine analogues as anti-malarial agents.

In the age of data-driven decision-making, database access is crucial for both non-technical and technical users. This study presents a powerful database chatbot that facilitates interaction with databases in human language and which does not require advanced SQL knowledge. It features chatbot-specific processes like handling user input, interpreting natural language, and presenting SQL queries; under the hood, this layer operates via Lang chain, where we leverage advanced language models. The chatbot can respond in a range of languages, catering to users who speak different languages.

Its multilingual support is one of the key reasons why this chatbot stands out, as it encourages users to engage in both regional and international languages, thereby including the entire spectrum of the population. Other important features are the inclusion of Google Text-to-Speech (GTTS), which makes this software text-to-speech- accessible, especially for users who have disabilities and want audio output. The app also allows users to copy responses to the clipboard and download all responses for greater flexibility and convenience.

Another advantage is session persistence. The chatbot can store session information so that it remembers the context of the conversation (i.e., I have been chatting with you and keep track of my previous messages, etc.). This is also powered by SQL and database semanticqueries, as well as context-aware responses to give better solutions. Future work will require further database compatibility, more query optimization, and advanced contextual conversation management to provide an even richer user experience.

As it bridges the gap between users and database systems using natural language processing, this project simplifies the way a database is handled by everyone over the globe, thereby making access to data easier through enhancing its usability for every kind of audience.

Based on the concepts of a circular economy and sustainability, this research aims to enrich foods by incorporating extracts from food by-products into them.

As the south of Galicia and the north of Portugal are major producers of Castanea sativa Miller chestnuts, it is imperative to devise sustainable uses for the by-products generated during their processing.

Chestnuts' consumption is widely promoted for its nutritional value and health benefits. It is a starchy fruit that is rich in vitamin C and minerals and gluten-free, making it great for people with coeliac disease, as well as being low in fat. There are several studies that describe the shells and hedgehogs of chestnuts as promising sources of bioactive compounds and fibre. Taking the nutritional advantages associated with both chestnut flour and chestnut by-products by various authors as a starting point, the aim is to extract tannins from the shells and hedgehogs for future incorporation into the flour. The shells will be heat-treated, dried, roasted, and boiled in thermal water. The hedgehogs will only be freeze-dried before being extracted. It is also hoped that we will ascertain the feasibility of incorporating the extracts of these by-products into chestnut flour, which is also made using different thermal processes, including drying, roasting, and boiling in thermal water, and is rich in minerals. The main aim of this study is to optimise a food product, chestnut flour, with added value by incorporating extracts from chestnut shells and hedgehogs. The aim is to produce a nutritionally enriched product with greater antioxidant capacity, thereby increasing the shelf life of chestnut flour.

The high price volatility of agri-horticultural commodities, especially essential ones like pulses, onions, and potatoes, poses formidable economic challenges to consumers, producers, and government bodies. The Department of Consumer Affairs is under constant pressure to stabilize markets, address inflation, and support sustainable agriculture. Farm Forecast solves this by providing a predictive analytics platform that combines artificial intelligence with cutting-edge algorithms to predict the future price movement of major agricultural commodities.

The Aurelius Market Intelligence System utilizes the ARIMA (Autoregressive Integrated Moving Average) model to examine past data and factors like weather, political affairs, supply chain bottlenecks, and trade scenarios globally. Farm Forecast is able to deliver an out-of-the-box forecast, stretching up to a full year ahead, on expected prices for maize, thus empowering different stakeholders like farmers, policy makers, and retailers to take informed decisions pertaining to production schedules, buffer stock management, pricing strategies, etc.

This user-friendly web dashboard offers interactive real-time insights and predictive analytics. This provides a chance for farmers to select crops as per the market demand, while government agencies can act before they are outsold due to sudden price changes. It also allows for automatic alerts to the Department of Consumer Affairs when market prices anticipate a shock (and hence increase uncertainty and price volatility).

In the future, Farm Forecast will be upgraded to include real-time data streams and operate in other crop markets. We plan to experiment with more advanced models like neural networks as the system matures to enhance prediction quality even further. Farm Forecast combines this mash-up of machine learning and economic modelling to provide a transformative tool for stabilizing commodity prices that benefits consumers as much as it does the agriculture industry.

Micro/nano-bubbles (MNBs) are tiny gas-filled cavities in bulk liquids. Recently, MNBs have drawn a lot of attention due to their industrial applications, such as in wastewater treatment, cleaning, disinfection, and other agriculture- and food-related applications due to their low cost, eco-friendliness, and ease of scale-up. Fermentation is an age-old practice adopted for the preservation of food and dairy products. Stirred yoghurt is one such product prepared by adding yoghurt starter culture and breaking the clumps after incubation. The present study attempted to understand the impact of the incorporation of MNBs on the fermentation pattern of yoghurt starter culture (YSC), i.e., Streptococcus thermophilus and Lactobacillus bulgaricus, and its rheological attributes when thus prepared. Different types of MNBs were prepared using compressed purified air, CO₂, O_2, and N₂ gas in water and milk systems using a bulk nanobubble generator. It was observed that the types of MNBs had a significant impact on the metabolism and microbial growth of the starter culture. Among the four different types of MNBs, the CO₂-MNBs had a significantly positive effect on bacterial growth besides increasing the viability of the fermented milk. Our findings suggest that MNBs in general and CO₂-MNBs specifically have the potential for applications in altering fermentation patterns. Furthermore, concerning the quality attributes of the MNBs incorporated into the stirred yoghurt, notable changes were observed in terms of its viscosity, mouthfeel, and shelf life. A significant rise in the viscosity of the stirred yoghurt with MNBs incorporated was observed as compared to that of the control sample, which may be attributed to the milk protein–polysaccharide interaction at the interface of the MNBs. It is therefore concluded that MNB technology has the potential to be used as a new processing tool to easily adjust the fermentation pattern and the rheological properties of fermented dairy products to fulfil growing consumer demand for innovative products with adjustable consistency and functionality.

Context: The productivity of agriculture is fully depending on the health and nutrient status of the soil. As the global population is increasing, the demand for food is increasing simultaneously. Traditional Farming practices always rely on general recommendation for soil management and crop selections. This may not be sufficient for different soil types and environmental conditions.

Objective: In this paper, we used the concept of the oversampling and under sampling techniques to balance the data. By using Machine Learning techniques, We analysed different types of soil nutrients for crops. We collected data on various attributes, like pH, Potassium(K), Rainfall, humidity, labels (different crops like rice, maize, etc.), temperature, etc. We used various Supervised Learning Algorithms in this project to predict the most efficient crop for different types of soil and environments like Decision Tree, Logistic Regression, SVM, Naive Byes, Random Forest, etc.

Method / Materials: We collected data from different sources and IoT sensors, like soil moisture sensors, pH sensors, temperature sensors, EC (Electrical Conductivity) sensors, nutrient sensors, etc). The collected dataset was converted into a .csv file. Further, we sent these data to the cloud over Wi-Fi. After analysing our data, we found that our dataset was unbalanced. So, we applied many data balancing techniques to balance our data for crop prediction. We have applied under sampling (counter) and oversampling (SMOTE) to our dataset.

Result: We have compared both under sampling and oversampling algorithms. After applying each of the algorithms, we obtained the best accuracy for Gaussian Naïve Bayes, which was 99.77% for crop recommendation and soil nutrient analysis. We estimated the performance metrics for all the classifiers and found that an accuracy of GNB 99.77% in comparison to the others.

Introduction: Software estimation plays a crucial role in predicting the budget and duration of software projects, and thus in their success. To estimate software, historical data are generally used, and machine learning is leveraged to predict software estimation. Vendors and companies share their past historical estimations with a central server, where a model is trained to evaluate and predict the estimation. However, all companies maintain confidential past estimations on their local servers. Objective: To eliminate this challenge, we use a federated learning framework to leverage AI and keep vendors' past estimates private in their local servers without sharing them with the central server. We also use machine learning to predict the software cost and duration. Material/Methods: In this framework, we train and evaluate the model in local servers and share their encrypted weights, bias, and accuracy score with a central server while ensuring past estimations are kept private. The central server will aggregate and average the weights and bias and share them back with the local server for retraining. This process will continue until it reaches the accuracy score necessary to predict and share the size of the new requirement with the central server and return the cost and duration. In the central server, we use a federated averaging algorithm for model aggregation, where the global model is updated by averaging the local model updates.

Result: The experimental results demonstrate the effectiveness of the proposed approach in accurate cost estimation and duration prediction by FLML. To evaluate the proposed framework, we conduct experiments using sample data. We compare the performance of the FL-based models with centralized models in terms of evaluation metrics. The performance will need to be improved (because of distributed/parallel data processing), but the FL framework provides privacy guarantees.

The accurate classification of gliomas from magnetic resonance (MR) imaging is vital for effective treatment planning. However, due to the irregular and diffuse boundaries of gliomas, manual classification is both difficult and time-consuming. To address these challenges, we present a novel methodology that combines image decomposition with local texture feature extraction to improve classification accuracy. The process begins by applying a Gaussian filter (GF) to the MR images to smooth them and reduce noise. Subsequently, Non-subsampled Laplacian Pyramid (NSLP) decomposition is used to capture multi-scale image details, which enhances the visibility of glioma boundaries. After decomposition, Total Variation-L1 (TV-L1) normalization is applied to reduce intensity inconsistencies, and Local Binary Patterns (LBPs) are utilized to extract key texture features from the processed images. These extracted texture features are then input into several supervised machine learning classifiers, including Support Vector Machines (SVMs), K-nearest Neighbors (KNNs), Decision Trees (DTs), AdaBoost, and LogitBoost. These models are trained to distinguish between low-grade (LG) and high-grade (HG) gliomas. Experimental results demonstrate that our proposed method consistently outperforms current state-of-the-art techniques in glioma classification, delivering superior accuracy in differentiating between LG and HG gliomas. This approach offers significant potential for improving diagnostic accuracy, thereby supporting clinicians in making informed and effective treatment decisions.

Speech emotion recognition (SER) is a particularly challenging task due to the intricate and non-linear features of emotional expressions in audio signals. In this work, we introduce SpectoResNet, an improved version of ResNet architecture that was tuned for classifying emotions using audio features from the CREMA-D dataset. CREMA-D, provided by the Speech and Emotion Research Group from New York University (NYU) , is a crowdsourced dataset consisting of 7,442 audio-visual recordings from 91 actors, which displays happiness, sadness, anger, and neutrality, among other emotions. While this dataset tries to provide opportunities for research into emotional recognition, its intrinsic variety and subtle differences due to individual traits and contextual environments pose significant obstacles to precise classification. To do this, we converted voice signals into 2D spectrograms to enable the deep CNN of ResNet to analyze and classify the emotions. ResNet was initially developed for image recognition and relies on residual connections in order to be able to train very deep networks effectively. Advanced data augmentation-adding noise and changing pitch-was used to simulate the variability found in real-time speech and make the model robust for different acoustic environments. Our model, trained on augmented spectrogram data, achieved 65.20% classification accuracy-a state-of-the-art breakthrough in vocal emotion recognition using deep learning. Success with SpectoResNet emphasizes the prowess of deep CNNs in extracting detailed patterns and subtleties within emotional audio expressions, thus paving the path toward more advanced model developments for multimodal emotion recognition.

Malaria, a life-threatening disease caused by Plasmodium parasites, remains a major global health challenge, particularly in regions with limited access to medical resources. Traditional diagnostic methods, such as microscopic examination of blood smears, are labor-intensive, time-consuming, and susceptible to human error, leading to delays in diagnosis and treatment. To overcome these challenges, we propose an automated system leveraging advanced computer vision and deep learning techniques, specifically utilizing the region-based fully convolutional neural network (R-FCN) object detection model. The R-FCN model is particularly adept at identifying and localizing objects within images, making it highly suitable for the accurate detection and classification of malaria parasites. Our system is trained on a labeled dataset of approximately 1,328 images, each annotated with bounding boxes to highlight the presence of malaria parasites. Through rigorous experimentation, our proposed system has demonstrated superior performance to baseline methods, achieving higher accuracy and efficiency in parasite detection. By automating the diagnostic process, our system significantly reduces the need for human intervention, thereby minimizing errors, accelerating diagnosis, and improving patient outcomes. Moreover, this approach holds great promise for streamlining the malaria diagnosis and treatment process globally, contributing to a broader effort to combat this devastating disease and enhance public health outcomes in affected regions.