Deafness and voice impairment have been persistent disabilities throughout history, hindering individuals from engaging in verbal communication and leading to their isolation from the predominantly vocally communicati...
详细信息
Various computer vision techniques based on deep neural networks have been proposed to detect objects accurately and fast. However, due to the privacy, security and communication bandwidth restrictions of diverse part...
详细信息
ISBN:
(数字)9781665490627
ISBN:
(纸本)9781665490627
Various computer vision techniques based on deep neural networks have been proposed to detect objects accurately and fast. However, due to the privacy, security and communication bandwidth restrictions of diverse participating parties, it is sometimes prohibitive to train such models on a centralized machine. Federated learning (FL) provides a promising solution to learn a model from decentralized data. Despite the advances in FL, the diversity of client regions in which they operate and the Non-IID nature of the crowdsourced datasets reduces the accuracy of object detection models significantly. In this paper, we introduce a novel FL object detection system to efficiently train models with heterogeneous client datasets. We propose lightweight client selection methods to learn object detection models faster. Our client selection methods based on the object data distribution at clients achieves up to 74% reduction in required federated rounds compared to conventional approaches. We further extend this method by leveraging the metadata of the training images (e.g., location, direction, depth), to select clients which maximize the coverage of diverse geographical regions. We report on extensive experiments with real datasets.
this study employs candlestick patterns and sequence similarities to forecast trends in time-series stock data. Time-series financial forecasting is a crucial component in risk management and portfolio optimization. F...
详细信息
this paper is devoted to the implementation of a modular case-based reasoning system for datamining based on previous experience in the form of cases. the case-based reasoning system allows you to work with cases pre...
详细信息
Feature selection in text classification refers to the critical process of identifying and selecting the most relevant and informative features such as words, phrases, or other linguistic elements from a text dataset....
详细信息
ISBN:
(纸本)9783031821523;9783031821530
Feature selection in text classification refers to the critical process of identifying and selecting the most relevant and informative features such as words, phrases, or other linguistic elements from a text dataset. this process, which has been a research topic for decades and finds applications across various fields such as bioinformatics, image recognition, image retrieval, text mining, etc., is essential for optimizing classification accuracy and efficiency. Addressing the challenge of high dimensionality in text data, stemming from the abundance of features like words or n-grams, is crucial to mitigate computational inefficiency and overfitting. Furthermore, the presence of irrelevant or redundant features in text datasets poses another significant challenge, as these features can introduce noise or irrelevant information, thereby underminingthe performance of classifier. In this paper, we proposed a new approach GL-SMCHI using improved features selection method to reduce the CHI value of high-frequency words, and globalization technique to incorporate both feature and class information when evaluating the importance of each feature. We compared the results obtained from our proposal against those of existing robust alternatives, the simulation results show that the proposed method outperforms the standard CHI squared, the improved feature selection methods, and the globalization method in term of F-Score and accuracy.
Open-set text recognition, which aims to address both novel characters and previously seen ones, is one of the rising subtopics in the text recognition field. However, the current open-set text recognition solutions o...
详细信息
ISBN:
(纸本)9783031705489;9783031705496
Open-set text recognition, which aims to address both novel characters and previously seen ones, is one of the rising subtopics in the text recognition field. However, the current open-set text recognition solutions only focuses on horizontal text, which fail to model the real-life challenges posed by the variety of writing directions in real-world scene text. Multi-orientation text recognition, in general, faces challenges from diverse image aspect ratios, significant imbalance in data amount, and domain gaps between orientations. In this work, we first propose a Multi-Oriented Open-Set Text recognition task (MOOSTR) to model the challenges of both novel characters and writing direction variety. We then propose a Multi-Orientation Sharing Experts (MOoSE) framework as a strong baseline solution. MOoSE uses a mixture-of-experts scheme to alleviate the domain gaps between orientations, while exploiting common structural knowledge among experts to alleviate the data scarcity that some experts face. the proposed MOoSE framework is validated by ablative experiments, and also tested for feasibility on an existing open-set text recognition benchmark. Code, models, and documents are available at: https://***/lancercat/Moose/
Twitter is the widely used microblogging site, where millions of people share their feelings, views, or opinion regarding different things be it a product, service, or events. Huge volumes of data are being produced h...
详细信息
Sensor-based human activity recognition (S-HAR) is a famous study focusing on detecting human physiological actions by interpreting various sensors, especially one-dimensional time series information. Typically, S-HAR...
详细信息
Preventing running injuries is critical for track and field athletes. this work presents a novel machinelearning methodology to automatically detect injury risk from videos by analyzing biomechanical technique. Joint...
详细信息
Identifying anomalies has become one of the primary strategies towards security and protection procedures in computer networks. In this context, machinelearning-based methods emerge as an elegant solution to identify...
详细信息
ISBN:
(数字)9781665490627
ISBN:
(纸本)9781665490627
Identifying anomalies has become one of the primary strategies towards security and protection procedures in computer networks. In this context, machinelearning-based methods emerge as an elegant solution to identify such scenarios and learn irrelevant information so that a reduction in the identification time and possible gain in accuracy can be obtained. this paper proposes a novel feature selection approach called Finite Element machines for Feature Selection (FEMa-FS), which uses the framework of finite elements to identify the most relevant information from a given dataset. Although FEMa-FS can be applied to any application domain, it has been evaluated in the context of anomaly detection in computer networks. the outcomes over two datasets showed promising results.
暂无评论