data Reduction without the removal of exact, correct rows is a crucial pre-processing step. Large datasets make it difficult to model data effectively or forecast results accurately. Additionally, they demand lengthy ...
data Reduction without the removal of exact, correct rows is a crucial pre-processing step. Large datasets make it difficult to model data effectively or forecast results accurately. Additionally, they demand lengthy processing times, sophisticated complexity software and thorough data cleaning. The incorrect and irrelevant rows may produce inaccurate results that impair the performance of the model. For better and more accurate outcomes it is crucial to properly detect and remove inaccurate data. The proposed algorithm calculates the Initial Recall value of the dataset. It eliminates least correlated features using the Correlation Matrix. Using Gaussian Curve, for all the columns it identifies and eliminates rows having values which lie beyond (μ ± 3σ). Furthermore, it takes into account the column with the highest Standard Deviation, selects the nearest 50% Left and 50% Right values from that column's mean. It selects only those rows and calculates the Final Recall value. Negligible difference between Initial and Final Recall values implies that the removed rows had no or minimal impact on the dataset's final result. This algorithm is implemented on 3 Standard Medical datasets - Pima Diabetes, Heart Attack and Breast Cancer. For the Breast Cancer dataset, this algorithm eliminated the highest number of rows that is 231.
Online advertising is the most popular and significant marketing tool in today's world. Traditional advertising systems failed to meet the needs of varied users, since they display ads without considering the indi...
详细信息
Online advertising is the most popular and significant marketing tool in today's world. Traditional advertising systems failed to meet the needs of varied users, since they display ads without considering the individual's characteristics. Now-a-days ads are displayed based on the previous search history of activities of the users, but, human mind change most frequently and the ads are being displayed regardless of the user's emotion. In this paper, we introduce an improved advertising system that predicts the user's sentiment based on the genre of the content being viewed and queues ads accordingly that uses both Collaborative and Content based Filtering that paved way for Hybrid Recommender System. This will enhance the user experience and the quality of the advertisements and advertising will be improved.
Lung cancer has become one of the primary causes of death worldwide due to cancer, which indicates the necessity of early diagnosis. In this regard, the present work proposes an optimized model of transfer learning ba...
详细信息
ISBN:
(数字)9798331518097
ISBN:
(纸本)9798331518103
Lung cancer has become one of the primary causes of death worldwide due to cancer, which indicates the necessity of early diagnosis. In this regard, the present work proposes an optimized model of transfer learning based on characterizing different stages of lung cancer by using computed tomography images acquired from the IQ-OTH/NCCD lung cancer database downloadable from Kaggle. It makes use of a two-step process: first, a U-Net model is used on the patient's CT images to reconstruct anatomical features and reduce noise; second, a ShuffleNet architecture is used to predict the patient's CT stage using the reconstructed image. In addition, to enhance classification, PSO was incorporated into the model so that it can optimize the hyperparameters. The accuracy of the model was at 97.85 percent, and it suggests that the model can become a very useful tool for clinicians to bring about earliest identification and staging for lung cancer.
Densely structured pruning methods utilizing simple pruning heuristics can deliver immediate compression and acceleration benefits with acceptable benign performances. However, empirical findings indicate such naï...
Densely structured pruning methods utilizing simple pruning heuristics can deliver immediate compression and acceleration benefits with acceptable benign performances. However, empirical findings indicate such naïvely pruned networks are extremely fragile under simple adversarial attacks. Naturally, we would be interested in knowing if such a phenomenon also holds for carefully designed modern structured pruning methods. If so, then to what extent is the severity? And what kind of remedies are available? Unfortunately, both questions remain largely unaddressed: no prior art is able to provide a thorough investigation on the adversarial performance of modern structured pruning methods (spoiler: it is not good), yet the few works that attempt to provide mitigation often do so at various extra costs with only to-be-desired *** this work, we answer both questions by fairly and comprehensively investigating the adversarial performance of 10+ popular structured pruning methods. Solution-wise, we take advantage of Grouped Kernel Pruning (GKP)'s recent success in pushing densely structured pruning freedom to a more fine-grained level. By mixing up kernel smoothness — a classic robustness-related kernel-level metric — into a modified GKP procedure, we present a one-shot-post-train-weight-dependent GKP method capable of advancing SOTA performance on both the benign and adversarial scale, while requiring no extra (in fact, often less) cost than a standard pruning procedure. Please refer to our GitHub repository for code implementation, tool sharing, and model checkpoints.
Research on classifying chest CT scans as normal or abnormal using machine learning and deep learning has garnered significant attention. To address this, various feature selection (FS) methods are employed to reduce ...
详细信息
In this paper, we present recent approaches proposed to secure the Internet of Things (IoT) devices against malicious cyber attacks and malware. As IoT devices have limited computing, storage, processing and communica...
详细信息
In this paper, we present recent approaches proposed to secure the Internet of Things (IoT) devices against malicious cyber attacks and malware. As IoT devices have limited computing, storage, processing and communication capabilities, protection those devices by themselves is very challenging. Fog computing has been proposed to support resource constrained IoT devices to reduce delay caused by cloud computing. We also presented adopted ML models and datasets vs. targeted cyber attacks in a tabular form.
While there have been considerable advancements in machine learning driven by extensive datasets, a significant disparity still persists in the availability of data across various sources and populations. This inequal...
详细信息
Chatbots have become a trending topic with emerging platforms like ChatGPT, Gemini, and Copilot, for conversation assistance. Current chatbots mainly focus on the general public assuming a natural flow of conversation...
详细信息
ISBN:
(数字)9798331529048
ISBN:
(纸本)9798331529055
Chatbots have become a trending topic with emerging platforms like ChatGPT, Gemini, and Copilot, for conversation assistance. Current chatbots mainly focus on the general public assuming a natural flow of conversation. However, there is a need for a chatbot that supports people with various communication disabilities. This research fills this gap by offering a novel technique for a chatbot that assists people with Aphasia, a condition characterised by difficulties with language. We propose a multi-modal chatbot that is customised and designed to assist users with communication disabilities in navigating a website. Unlike typical chatbots, which rely on one form of communication, our architecture combines multiple modalities to improve comprehension and promote effective communication for people with Aphasia. We focus on gathering multimodal inputs by recognising and combining user intents from diverse sources. The use of Txtai, an all-in-one embeddings database for semantic search improves our chatbot’s capacity to process various inputs efficiently. We leverage specialised models like Whisper for audio transcription and MediaPipe Gesture Recognizer for gesture detection to enhance user interactions. Additionally, Rasa Core integration improves conversational experiences for users. We propose that this new approach will make communication more accessible and inclusive for individuals with Aphasia.
Automated data insight mining and visualization have been widely used in various business intelligence applications (e.g., market analysis and product promotion). However, automated insight mining techniques often out...
详细信息
In the continually evolving field of vehicular communications, the efficient allocation of time slots for Vehicle-to-Vehicle (V2V) communication is of utmost importance. This work introduces a novel approach that empl...
详细信息
ISBN:
(数字)9798350360790
ISBN:
(纸本)9798350360806
In the continually evolving field of vehicular communications, the efficient allocation of time slots for Vehicle-to-Vehicle (V2V) communication is of utmost importance. This work introduces a novel approach that employs a K-Nearest Neighbors (KNN) algorithm to allocate time slots in NR V2X sidelink communications. The allocation is specifically tailored to dynamically adjust to the current density of traffic in real-time. The efficacy of the KNN-based time slot allocation system is assessed through simulating the V2X environment, encompassing a fleet of vehicles. The efficacy of the suggested methodology is assessed in relation to a random allocation technique and it is observed that the suggested strategy notably enhances allocation efficiency, slot utilization, contention window, and communication time. The findings illustrate the capacity of a KNN-based dynamic allocation system to adapt to different levels of traffic congestion, leading to the efficient utilization of network resources and the mitigation of communication latency. Furthermore, the examination of the implementation of priority queues in the allocation process is explored, resulting in enhanced efficiency of the system in scenarios involving a substantial amount of traffic.
暂无评论