Theagarajan Padma

Proper waste management has been emergent attention in varied human habitats. Following the best practices for waste management is essential for a sustainable living environment. This research work proposes a technology-supported self-sustainable aquaponics environment that automatically manages and controls the system by integrating with IoT technology and Naive Bayes algorithm for automated fish feeding. Water quality is monitored with sensors such as pH, temperature, humidity, dissolved oxygen, and water level sensors. Solid waste of fish is filtered and the nitrification process has been carried out by bio-filter. The water level of the fish tank is monitored and maintained by an auto system. Fish feeding requirements have been attained using ML model. The nutrient film technique-based planting system optimally extracts nutrients. The Vertical farming technique helps to reduce the land and water used for cultivation. The sensors are operated on microcontrollers namely Arduino UNO and Raspberry Pi. The sensed values are communicated through a mobile application for constantly monitoring the aquaponics environment. The prospect shows the self-sustainable smart aquaponic platform for farmers to grow fish and plants in a single system simultaneously to get increased production with fewer natural resources such as land and water.

An early stage of breast cancer occurrence does not have pain as a symptom. This asymptomatic nature of cancer necessitates the need of timely and accurate prediction using other potential indicators. Breast cancers are highly curable if predicted and diagnosed at the earliest. This research explores the capacity of hybridizing intelligent learning models with nature-inspired optimization algorithms for breast cancer prediction. This integration becomes pivotal to optimize internal parameters of the dataset while aiming for augmented accuracy of the proposed predictive models. Each of the four intelligent machine learning models including K-Nearest Neighbor (KNN), Naïve Bayes (NB), Support Vector Machine (SVM), and Artificial Neural Network (ANN) are hybridized with all the three adaptive optimization approaches including Adaptive Particle Swarm Optimization (APSO), Adaptive Genetic Algorithm (AGA) and Adaptive Venus Flytrap Optimization (AVFO). Thus twelve hybridized predictive models were derived namely APSO-KNN, AGA-KNN, AVFO-KNN, APSO-NB, AGA-NB, AVFO-NB, APSO-SVM, AGA-SVM, AVFO-SVM, APSO-ANN, AGA-ANN and AVFO-ANN. These hybridized models were investigated through UCI data repository’s breast cancer dataset namely WBC, WDBC and WPBC. The experimental results were validated against the respective learning models with and without feature selection. Based on the performance measures such as accuracy, F-measure and G-mean, the outperformed learning models with WBC, WPBC and WDBC datasets are AVFO-SVM, AVFO-ANN and APSO-ANN respectively. Conclusively ANN and SVM machine learning algorithms fused with AVFO for feature selection are robust for all the three datasets. The derived hybrid intelligent models trained with tuned datasets optimize the prediction ability of existing breast cancer prediction models.

Agriculture has an important role in India’s economic development. Crop productivity is affected by the rising population and the country’s ever-changing climate. Crop yield estimation is a challenge in the farming sector. Numerous studies have been conducted in the agricultural sector to better estimate crop yield through machine learning techniques. It is an efficient method for anticipating crop yields and determining which crops to cultivate. Random Forest has been widely utilized for this purpose. A set of parameters in the Random Forest classifier must be stay tuned. The machine learning algorithm will yield better results with correct hyper parameter adjustment. This work presents a hybrid approach to agricultural yield estimation using a Random Forest classifier and the Random Search method with a 0.99 R2 score, 0.045 MSE, and 0.022 MAE, the suggested method outperformed other existing approaches such as Decision Tree (DT), Multiple Linear Regression (MLR), Random Forest (RF), and Grid Search (GS) optimized RF. Validation methods such as R2, Mean Squared Error, and Mean Absolute Error to cross-validation have been used to confirm the authenticity of the outcomes. The purpose of this study is to apply the crop yield prediction approach into action to assist farmers in solving agricultural production concerns.

Purpose: Conventional diagnostic techniques, on the other hand, may be prone to subjectivity since they depend on assessment of motions that are often subtle to individual eyes and hence hard to classify, potentially resulting in misdiagnosis. Meanwhile, early nonmotor signs of Parkinson’s disease (PD) can be mild and may be due to variety of other conditions. As a result, these signs are usually ignored, making early PD diagnosis difficult. Machine learning approaches for PD classification and healthy controls or individuals with similar medical symptoms have been introduced to solve these problems and to enhance the diagnostic and assessment processes of PD (like, movement disorders or other Parkinsonian syndromes).Design/methodology/approach: Medical observations and evaluation of medical symptoms, including characterization of a wide range of motor indications, are commonly used to diagnose PD. The quantity of the data being processed has grown in the last five years; feature selection has become a prerequisite before any classification. This study introduces a feature selection method based on the score-based artificial fish swarm algorithm (SAFSA) to overcome this issue.Findings: This study adds to the accuracy of PD identification by reducing the amount of chosen vocal features while to use the most recent and largest publicly accessible database. Feature subset selection in PD detection techniques starts by eliminating features that are not relevant or redundant. According to a few objective functions, features subset chosen should provide the best performance.Research limitations/implications: In many situations, this is an Nondeterministic Polynomial Time (NP-Hard) issue. This method enhances the PD detection rate by selecting the most essential features from the database. To begin, the data set's dimensionality is reduced using Singular Value Decomposition dimensionality technique. Next, Biogeography-Based Optimization (BBO) for feature selection; the weight value is a vital parameter for finding the best features in PD classification.Originality/value: PD classification is done by using ensemble learning classification approaches such as hybrid classifier of fuzzy K-nearest neighbor, kernel support vector machines, fuzzy convolutional neural network and random forest. The suggested classifiers are trained using data from UCI ML repository, and their results are verified using leave-one-person-out cross validation. The measures employed to assess the classifier efficiency include accuracy, F-measure, Matthews correlation coefficient.

Worldwide demand for fish products is increasing continuously. Literature evidence indicates that there is a persistent decrease in ocean fisheries’ supply. Aquaculture bridges the gap between the reduced ocean fish supply and increased world fish food demand. Sustainable and profitable aquaculture is firmly facilitated by selective fish species. Species selection has been achieved through deeply analyzing the manifold and complex interrelationships between the numerous subjective risk categories intricate in aquaculture. Apparently an analytical system able to analyze massive subjective stakes in terms of its quantifiable equivalent that aids selecting an optimal fish species is nontrivial. This research provides quantifiable metrics that eases the analytical struggle towards the subjective aspect of species selection. The novelty involves providing hybrid multi-criteria-based viable decision support methodology that analyses extensive aquaculture domain knowledge and inference and emulates the logic and reasoning process so as to choose an optimal fish species for aqua farming. The methodology is based on the assessment of several species in accordance with analyzing numerous associated criteria and sub-criteria of risk factors. This research consists of nineteen sub-criteria which were classified under five comprehensive heads of evaluation criteria such as environmental, nutritional, disease outbreaks, biotic and physiological risk categories. The weight scores for each criteria and sub-criteria were determined using the Fuzzy Analytical Hierarchy Processing (FAHP) method. Consequently using the derived priority weights, the best fish species to choose from several alternative species is identified based on the relative closeness values and ranks assigned to them by applying the fuzzy Technique for Order Preference by Similarity to the Ideal Solution (TOPSIS) method. A case study is performed on five varieties of most favored fish species for consumption to exemplify the effectiveness of the proposed model. The sturdiness of the suggested model is validated against the two standing multiple criteria decision-making approaches such as fuzzy ordered weighted average and fuzzy extent analysis–fuzzy weighted average methods. The research outcome strongly aids aqua farmers to identify an optimal fish species.

Due to the drastic exploitation of mobile devices and mobile apps in the day-to-day activities of people, the enhancement in hardware and software tools for mobile devices is also rising rapidly to cater to the requirements of mobile users. However, the progress of resource-intensive mobile applications is still inhibited by the limited battery power, restricted memory, and scarce resources of mobile devices. By employing mobile cloud computing, mobile edge computing, and fog computing, many researchers are providing their frameworks and offloading algorithms to augment the resources of mobile devices. In the existing solutions, offloading resource-intensive tasks is adopted only for specific scenarios and also not supporting the flexible exploitation of IoT-based smart mobile applications. So, a novel neuro-fuzzy modeling framework is proposed to augment the inadequate resources of a mobile device by offloading the resource-intensive tasks to external entities, and also a Bat optimization algorithm is exploited to schedule as many tasks as possible to the augmentation entities thereby improving the total execution time of all tasks and minimizing the resource exploitation of the mobile device. In this research work, external augmentation entities like distant cloud, edge cloud, and microcontroller devices are providing Resource augmentation as a Service (RaaS) to mobile devices. An IoT-based smart transport mobile app is implemented based on the proposed framework which depicts a significant reduction in execution time, energy consumption, bandwidth utilization, and average delay. Performance analysis depicts that the neuro-fuzzy hybrid model with Bat optimization provides a significant improvement compared with proximate computing and web service frameworks on the Quality of Service (QoS) parameters namely energy consumption, execution time, bandwidth utilization, and latency. Thus, the proposed framework exhibits a feasible solution of RaaS to resource-constrained mobile devices by exploiting edge computing.