Browse Items (2150 total)
Sort by:
-
Machine Learning based Plant Disease Identification by using Hybrid Nae Bayes with Decision Tree Algorithm
Artificial intelligence or machine learning as a domain started as a distinct domain marketplace for enthusiasts. Over an extended period of time, this has evolved into an industry with boundless potential. This is the focal point of a plethora of technologies like real-time analytics, deep learning in computer science. It's inherent to various customer needs such as fault detection, home automation, health monitoring devices as well as appliances, and multiple RPM devices Artificial intelligence which has been tested and trained to recognize and determine a plethora of flaws and inaccuracies. This could be intriguing procedures in day-to-day applications. An unimaginable number of prediction models, packages, libraries as well as sensors are utilized to sieve through flaws with the aid of mobile app development and other multispectral sensors. These trendy devices have become ever present and a part of our extensive routine. The demand for dependable and efficient algorithms is satisfied while implementing these devices. The objective primarily dictates emphasis on the prediction of plant diseases in the agricultural arena in reality by providing aid in the field of agriculture, and industry. In this case, the device incorporates a database which stores and keeps track of previously detected flaws or defects. In addition, the history of detected plant infections is maintained in an online repository. This can help with the forecast of the defects within the gadgets that are to be enhanced. Furthermore, the suggested approach of this text inculcates the invigilation of every leaf in the plant via machine learning model. Hence, this approach of implementation limits interaction of humans with the interface and it detects disease ridden plants efficiently with accuracy. The plant disease identification problem is to solve the proposed hybrid Nae Bayes with Decision Tree algorithm. The proposed model provides higher accuracy level compare to the regular model. 2023 IEEE. -
Investigating Key Contributors to Hospital Appointment No-Shows Using Explainable AI
The healthcare sector has suffered from wastage of resources and poor service delivery due to the significant impact of appointment no-shows. To address this issue, this paper uses explainable artificial intelligence (XAI) to identify major predictors of no-show behaviours among patients. Six machine learning models were developed and evaluated on this task using Area Under the Precision-Recall Curve (AUC-PR) and F1-score as metrics. Our experiment demonstrates that Support Vector Classifier and Multilayer Perceptron perform the best, with both scoring the same AUC-PR of 0.56, but different F1-scores of 0.91 and 0.92, respectively. We analysed the interpretability of the models using Local Interpretable Model-agnostic Explanation (LIME) and SHapley Additive exPlanations (SHAP). The outcome of the analyses demonstrates that predictors such as the patients' history of missed appointments, the waiting time from scheduling time to the appointments, patients' age, and existing medical conditions such as diabetes and hypertension are essential flags for no-show behaviours. Following the insights gained from the analyses, this paper recommends interventions for addressing the issue of medical appointment no-shows. 2024 IEEE. -
Parametric Study on Compaction Characteristics of Clay Sand Mixtures
The behaviour of fine-grained soils can be attributed to their mineral composition and the amount of fines present in them. The present study aims to determine the effect of mineral composition and quantity of fines on the Atterberg limits and compaction characteristics and to determine the correlation between them. Two types of fine-grained artificial soil mixtures were prepared in the laboratory representing kaolinitic and montmorillonitic mineral compositions.The amount of fines was varied at 10% intervals, from 50 to 100%. The Atterberg limits like liquid limit, plastic limit, shrinkage limit, and compaction characteristics like maximum dry density (MDD) and optimum moisture content (OMC) for two compaction energy levels, i.e. standard proctor (SP) and modified proctor (MP) tests, were determined. The correlations were developed between percentage fines and Atterberg limits and similarly between percentage fines, Atterberg limits, and compaction characteristics for artificial mix proportions. The developed correlations were used to predict the properties of natural soil samples, and the predicted and actual values are compared. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Text Summarization Techniques for Kannada Language
Text Summarization is summarizing the original text document into a shorter description. This short version should retain the meaning and information content of the original text document. A concise summary can help humans quickly understand a large original document better in a short time. Summarization can be used in many text documents, such as reviews of books, movies, newspaper articles, content, and huge documents. Text summarization is broadly classified into extractive Text Summarization (ETS) and Abstractive Text Summarization (ATS). Even though more research works are carried out using extractive methods, meaningful summaries can be attained using abstractive summary techniques, which are more complex. In Indian languages, very few works are carried out in abstract summarization, and there is a high need for research in this area. The paper aims to generate extractive and abstractive summaries of the text by using deep learning and extractive summaries and comparisons between them in the Kannada language. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
Perception to Control: End-to-End Autonomous Driving Systems
End-to-end autonomous driving systems have garnered a lot of attention in recent years, and researchers have been exploring different ways to make them work. In this paper, we provide an overview of the field with a focus on the two main types of systems: those that use only RGB images and those that use a combination of multiple modalities. We review the literature in each area, highlighting the strengths and limitations of each approach. We also discuss the challenges of integrating these systems into a complete end-to-end autonomous driving pipeline, including issues related to perception, decision-making, and control. Lastly, we identify areas where more research is needed to make autonomous driving systems work better and be safer. Overall, this paper provides a comprehensive look at the current state-of-the-art in end-to-end autonomous driving, with a focus on the technical challenges and opportunities for future research. The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024. -
X-Tract: Framework for Flexible Extraction of Features in Chest Radiographs for Disease Diagnosis Using Machine Learning
Various types of medical images are used as diagnostic tools for identifying pathologies in human bodies, and in this research, chest X-ray images are used as diagnostic tools. Several pre-built models are created by the participants of ImageNet competitions for non-medical images, and these models are also being used in medical image classification; for example, Khan et al. (Comput Methods Prog Biomed 196:105581, 2020) developed a model called Coronet and Narayan Das et al. (IRBM 1:16, 2020) proposed a deep transfer learning-based model. Instead of using the pre-built models, a different approach was taken to address this problem. A framework was created to extract the frequency and spatial domain-based features, along with the raw statistics of the images. The model proposed in this article using the SVM algorithm has reached accuracy levels ranging from 91% to 97% and sensitivity of 92% to 96% on various samples of test data of over 400 images. 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Hyperspectral Image Classification Using Denoised Stacked Auto Encoder-Based Restricted Boltzmann Machine Classifier
This paper proposes a novel solution using an improved Stacked Auto Encoder (SAE) to deal with the problem of parametric instability associated with the classification of hyperspectral images from an extensive training set. The improved SAE reduces classification errors and discrepancies present within the individual classes. The data augmentation process resolves such constraints, where several images are produced during training by adding noises with various noise levels over an input HSI image. Further, this helps in increasing the difference between multiple classes of a training set. The improved SAE classifies HSI images using the principle of Denoising via Restricted Boltzmann Machine (RBM). This model ambiguously operates on selected bands through various band selection models. Such pre-processing, i.e., band selection, enables the classifier to eliminate noise from these bands to produce higher accuracy results. The simulation is conducted in PyTorch to validate the proposed deep DSAE-RBM under different noisy environments with various noise levels. The simulation results show that the proposed deep DSAE-RBM achieves a maximal classification rate of 92.62% without noise and 77.47% in the presence of noise. 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG. -
Firefly Algorithm andDeep Neural Network Approach forIntrusion Detection
Metaheuristic optimization has grown in popularity as a way for solving complex issues that are difficult to solve using traditional methods. With fast growth of the available storage space and processing capabilities of the modern computers, the machine learning domain, that can be succinctly formulated as the process of enabling the computers to make successful forecasts based on the previous experiences, has recently been under spectacular growth. This paper presents intrusion detection approach by utilizing hybrid method between firefly algorithm and deep neural network. The basic firefly algorithm, as a frequently employed swarm intelligence method, has several known deficiencies, and to overcome them, an enhanced firefly algorithm was proposed and used in this manuscript. For experimental purposes, KDD Cup 99 and NSL-KDD datasets from Kaggle and UCL repositories were taken and comparison with other frameworks that have been validated for the same datasets was executed. Based on simulation data, proposed method was able to establish better values for accuracy, precision, recall, F-score, sensitivity and specificity metrics than other approaches. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. -
Chaotic Binary Ant Lion Optimizer Approach for Feature Selection on Medical Datasets with COVID-19 Case Study
Binary version of the ant lion optimizer (ALO) are suggested and utilized in wrapper-mode to pick the best feature subset for classification. ALO is a recently developed bio-inspired optimization approach that mimics ant lion hunting behavior. Furthermore, ALO balances exploration and exploitation utilizing a unique operator to explore the space of solutions adaptively for the best solution. The difficulties of a plethora of noisy, irrelevant, and misleading features, as well as the capacity to deal with incorrect and inconsistent data in real-world subjects, provide rationale for feature selection to become one of the most important requirements. A difficult machine learning problem is to choose a subset of important characteristics from a vast number of features that characterize a dataset. Choosing the most informative markers and conducting a high-accuracy classification across the data may be a difficult process, especially if the data is complex. The feature selection task is usually expressed as a bio-objective optimization challenge, with the goal of enhancing the performance of the prediction model (data training fitting quality) while decreasing the number of features used. Various evaluation criteria are employed to determine the success of the suggested approach. The findings show that the suggested chaotic binary algorithm can explore the feature space for optimum feature set efficiently. 2022 IEEE. -
Assessing Player Interaction for a Social Networking Cooperative Educational Game
Cooperative interaction in educational games, designed to stimulate teamwork, joint creativity and knowledge sharing, also carries potential security threats. One of the key dangers is data leakage. Player interaction involves the exchange of information, and in case of insufficient protection of the system, confidential data, such as personal information, game progress results or individual strategies, may become available to unauthorized persons. This may result in misuse of information, damage to reputation and violation of player privacy. The impact on the game space is also a threat. By interacting, players can change the game world, for example, by entering incorrect data, moving objects to an inappropriate location, or modifying the rules of the game. This can lead to a violation of the balance of the game, incorrect results and a deterioration in the learning effect. Substitution or falsification of game elements is no less dangerous. Attackers can introduce fake elements into the game space, for example, incorrect reviews, changed rules or incorrect data. This can lead to incorrect conclusions, distort learning outcomes, and undermine confidence in the game. In addition, the use of interaction tools can become an object of attack. Attackers can hack and modify tools, such as communication platforms or data storage systems. This can lead to data theft, incorrect operation of tools and malfunction during the game. It is shown that formal descriptions of the choice of a game strategy can exist in a game. Indicators that are essential for cooperative interaction are determined, and examples of their calculation for the case with remote interaction through a social network are given. The article contains information about collaborations, which can be used to assess and choose the direction of development in projects that use game cooperative strategies to implement tasks other than training. The project highlights aspects of cooperative interaction that affect the formation of game strategies in an educational project. Of particular interest are projects in which a social network is the tool and medium of interaction. The objectives of the project are to identify easy-to-use indicators that show the features of cooperative interaction within an educational game. The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.