检索结果-内蒙古大学图书馆

Value Identification in Multistakeholder Recommender Systems for Humanities and Historical Research: The Case of the Digital Archive ***

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Atzenhofer-Baumgartner, Florian Geiger, Bernhard C. Vogeler, Georg Kowald, Dominik Institute of Interactive Systems and Data Science Graz University of Technology Graz Austria Department of Digital Humanities University of Graz Graz Austria Know Center Research GmbH Graz Austria Signal Processing and Speech Communication Laboratory Graz University of Technology Graz Austria

Recommender systems remain underutilized in humanities and historical research, despite their potential to enhance the discovery of cultural records. This paper offers an initial value identification of the multiple stakeholders that might be impacted by recommendations in ***, a digital archive for historical legal documents. Specifically, we discuss the diverse values and objectives of its stakeholders, such as editors, aggregators, platform owners, researchers, publishers, and funding agencies. These in-depth insights into the potentially conflicting values of stakeholder groups allow designing and adapting recommender systems to enhance their usefulness for humanities and historical research. Additionally, our findings will support deeper engagement with additional stakeholders to refine value models and evaluation metrics for recommender systems in the given domains. Our conclusions are embedded in and applicable to other digital archives and a broader cultural heritage context. © 2024, CC BY-SA.

关键词： Embedded systems

C2P-CLIP: Injecting Category Common Prompt in CLIP to Enhance Generalization in Deepfake Detection 39

学校读者我要写书评

暂无评论

C2P-CLIP: Injecting Category Common Prompt in CLIP to Enhanc...

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Tan, Chuangchuang Tao, Renshuai Liu, Huan Gu, Guanghua Wu, Baoyuan Zhao, Yao Wei, Yunchao Institute of Information Science Beijing Jiaotong University China Visual Intellgence +X International Cooperation Joint Laboratory of MOE China Pengcheng Laboratory Shenzhen China School of Information Science and Engineering Yanshan University China Hebei Key Laboratory of Information Transmission and Signal Processing China School of Data Science The Chinese Univ of Hong Kong CUHK-Shenzhen Shenzhen China

ISBN: (纸本)157735897X

This work focuses on AIGC detection to develop universal detectors capable of identifying various types of forgery images. Recent studies have found large pre-trained models, such as CLIP, are effective for generalizable deepfake detection along with linear classifiers. However, two critical issues remain unresolved: 1) understanding why CLIP features are effective on deepfake detection through a linear classifier;and 2) exploring the detection potential of CLIP. In this study, we delve into the underlying mechanisms of CLIP’s detection capabilities by decoding its detection features into text and performing word frequency analysis. Our finding indicates that CLIP detects deepfakes by recognizing similar concepts. Building on this insight, we introduce Category Common Prompt CLIP, called C2P-CLIP, which integrates the category common prompt into the text encoder to inject category-related concepts into the image encoder, thereby enhancing detection performance. Our method achieves a 12.4% improvement in detection accuracy compared to the original CLIP. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词：

Improving Mandarin End-to-End Speech Recognition with Word N-gram Language Model

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Tian, Jinchuan Yu, Jianwei Weng, Chao Zou, Yuexian Yu, Dong The Advanced Data and Signal Processing Laboratory School of Electric and Computer Science Peking University Shenzhen Graduate School Shenzhen China Tencent AI Lab Tencent ASR Oteam

Despite the rapid progress of end-to-end (E2E) automatic speech recognition (ASR), it has been shown that incorporating external language models (LMs) into the decoding can further improve the recognition performance of E2E ASR systems. To align with the modeling units adopted in E2E ASR systems, subword-level (e.g., characters, BPE) LMs are usually used to cooperate with current E2E ASR systems. However, the use of subword-level LMs will ignore the word-level information, which may limit the strength of the external LMs in E2E ASR. Although several methods have been proposed to incorporate word-level external LMs in E2E ASR, these methods are mainly designed for languages with clear word boundaries such as English and cannot be directly applied to languages like Mandarin, in which each character sequence can have multiple corresponding word sequences. To this end, we propose a novel decoding algorithm where a word-level lattice is constructed on-the-fly to consider all possible word sequences for each partial hypothesis. Then, the LM score of the hypothesis is obtained by intersecting the generated lattice with an external word N-gram LM. The proposed method is examined on both Attention-based Encoder-Decoder (AED) and Neural Transducer (NT) frameworks. Experiments suggest that our method consistently outperforms subword-level LMs, including N-gram LM and neural network LM. We achieve state-of-the-art results on both Aishell-1 (CER 4.18%) and Aishell-2 (CER 5.06%) datasets and reduce CER by 14.8% relatively on a 21K-hour Mandarin dataset. © 2022, CC BY-NC-SA.

关键词： Speech recognition

A Joint Learning Sentiment Analysis Method Incorporating Emoji-Augmentation

学校读者我要写书评

暂无评论

A Joint Learning Sentiment Analysis Method Incorporating Emo...

IEEE International Conference on Cloud Computing and Intelligence Systems (CCIS)

作者： Jie Chen Luping Luo Bojing Ji Shu Zhao Yanping Zhang The Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Computer Science and Technology Anhui University China School of Big Data and Artificial Intelligence Chizhou University China

ISBN: (纸本)9781665456579

Social media is the platform for most people to share their opinions, emojis are also widely used to express moods, emotions, and feelings on social media. There have been many researched on emojis and sentiment analysis. However, existing methods mainly face two limitations. First, since deep learning relies on large amounts of labeled data, the training samples of emoji are not enough to achieve the training effect. Second, they consider the sentiment of emojis and texts separately, not fully exploring the impact of emojis on the sentiment polarity of texts. In this paper, we propose a joint learning sentiment analysis method incorporating emoji-augmentation, and the method has two advantages compared with the existing work. First, We optimize the easy data augmentation method so that the newly generated sentences can also preserve the semantic information of emojis, which relieves the problem of insufficient training data with emojis. Second, it fuses emojis and text features to allow the model to better learn the mutual emotional semantics between text and emojis, jointly training emojis and words to obtain the sentence representations containing more semantic information of both emojis and text. Our experimental results show that the proposed method can significantly improve the performance compared with several baselines on two datasets.

关键词： Training Sentiment analysis Social networking (online) Fuses Text recognition Mood Semantics

Integrating Lattice-Free MMI into End-to-End Speech Recognition

学校读者我要写书评

暂无评论

arXiv 2022年

In automatic speech recognition (ASR) research, discriminative criteria have achieved superior performance in DNN-HMM systems. Given this success, the adoption of discriminative criteria is promising to boost the performance of end-to-end (E2E) ASR systems. With this motivation, previous works have introduced the minimum Bayesian risk (MBR, one of the discriminative criteria) into E2E ASR systems. However, the effectiveness and efficiency of the MBR-based methods are compromised: the MBR criterion is only used in system training, which creates a mismatch between training and decoding;the on-the-fly decoding process in MBR-based methods results in the need for pre-trained models and slow training speeds. To this end, novel algorithms are proposed in this work to integrate another widely used discriminative criterion, lattice-free maximum mutual information (LF-MMI), into E2E ASR systems not only in the training stage but also in the decoding process. The proposed LF-MMI training and decoding methods show their effectiveness on two widely used E2E frameworks: Attention-Based Encoder-Decoders (AEDs) and Neural Transducers (NTs). Compared with MBR-based methods, the proposed LF-MMI method: maintains the consistency between training and decoding;eschews the on-the-fly decoding process;trains from randomly initialized models with superior training efficiency. Experiments suggest that the LF-MMI method outperforms its MBR counterparts and consistently leads to statistically significant performance improvements on various frameworks and datasets from 30 hours to 14.3k hours. The proposed method achieves state-of-the-art (SOTA) results on Aishell-1 (CER 4.10%) and Aishell-2 (CER 5.02%) datasets. Code is released. © 2022, CC BY-NC-ND.

关键词： Decoding

CHALLENGES IN IMPLEMENTING A RECOMMENDER SYSTEM FOR HISTORICAL RESEARCH IN THE HUMANITIES

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Atzenhofer-Baumgartner, Florian Geiger, Bernhard C. Trattner, Christoph Vogeler, Georg Kowald, Dominik Institute of Interactive Systems and Data Science Graz University of Technology Graz Austria Department of Digital Humanities University of Graz Graz Austria Know Center Research GmbH Graz Austria Signal Processing and Speech Communication Laboratory Graz University of Technology Graz Austria University of Bergen & Media Futures Bergen Norway

This extended abstract describes the challenges in implementing recommender systems for digital archives in the humanities, focusing on ***, a platform for historical legal documents. We discuss three key aspects: (i) the unique characteristics of so-called charters as items for recommendation, (ii) the complex multi-stakeholder environment, and (iii) the distinct information-seeking behavior of scholars in the humanities. By examining these factors, we aim to contribute to the development of more effective and tailored recommender systems for (digital) humanities research. © 2024, CC BY-SA.

关键词： Recommender systems

Automatic Detection and Characterization of Human Veins Using Infra-Red Image processing

学校读者我要写书评

暂无评论

Journal of Computer and Communications 2024年第9期12卷 141-159页

作者： Jean Ndoumbe Brice Ekobo Akoa Gaelle Patricia Talotsing Frederic Franck Kounga Samuel Kaissassou Bertin Chouanmo Njo Laboratory of Computer Engineering Data Science and Artificial Intelligence Department of Computer & Telecommunications Engineering National Higher Polytechnic School Douala Cameroon Laboratory of Electrical Engineering Mechatronics and Signal Processing Department of Electrical & Telecommunications Engineering National Advanced School of Engineering Yaound Cameroon Mobile Computing and Networking Research Laboratory Department of Computer and Software Engineering Polytechnique Montreal Montreal Canada DigiPlus SARL Bonamoussadi Douala Cameroon

The detection and characterization of human veins using infrared (IR) image processing have gained significant attention due to its potential applications in biometric identification, medical diagnostics, and vein-based authentication systems. This paper presents a low-cost approach for automatic detection and characterization of human veins from IR images. The proposed method uses image processing techniques including segmentation, feature extraction, and, pattern recognition algorithms. Initially, the IR images are preprocessed to enhance vein structures and reduce noise. Subsequently, a CLAHE algorithm is employed to extract vein regions based on their unique IR absorption properties. Features such as vein thickness, orientation, and branching patterns are extracted using mathematical morphology and directional filters. Finally, a classification framework is implemented to categorize veins and distinguish them from surrounding tissues or artifacts. A setup based on Raspberry Pi was used. Experimental results of IR images demonstrate the effectiveness and robustness of the proposed approach in accurately detecting and characterizing human. The developed system shows promising for integration into applications requiring reliable and secure identification based on vein patterns. Our work provides an effective and low-cost solution for nursing staff in low and middle-income countries to perform a safe and accurate venipuncture.

关键词： Vein Detection Blood Radiation Infrared Image CLAHE Algorithm Raspberry Pi

Dadm: Enhancing Eeg Artifact Removal by Fusing Artifact Representation into the Time-Frequency Domain

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Li, Haoran Feng, Fan Kang, Jiarong Zhang, Jin Gong, Xiaoli Lu, Tingjuan Sun, Zhe Solé-Casals, Jordi College of Computer Science Nankai University Tianjin China Department of Biomedical Engineering School of Medicine Tsinghua University Beijing China Tianjin Key Laboratory of Brain Science and Intelligent Rehabilitation Nankai University Tianjin China State Key Laboratory of High-end Server and Storage Technology China 903rd Hospital of PLA China Faculty of Medicine Faculty of Health Data Science Juntendo University Japan Data and Signal Processing Group University of Vic Central University of Catalonia Catalonia Vic08500 Spain

The electroencephalogram (EEG) provides essential data for analyzing brain activities. However, artifacts such as electrooculography (EOG) and electromyography (EMG) often interleave with the EEG signals, significantly affecting the quality of EEG signal analysis. The heterogeneous distribution of these artifacts in the time-frequency domain makes it challenging to remove multiple artifacts using a unified model. In this paper, we propose an Artifact-Aware EEG Denoising Model, referred to as DADM, to effectively remove various types of artifacts in a unified manner. We first obtain an artifact representation that indicates the type of artifact from a pre-trained artifact classification model. This artifact representation is then used as prior knowledge, which is fused into the denoising model in the frequency-time domain. This enables the model to become aware of the artifact type and precisely remove artifacts based on their type. Due to the heterogeneous distributions of artifacts in the frequency domain, we introduce a frequency enhancement module that can identify specific types of artifacts based on their representation and remove them using a hard attention mechanism. Additionally, we design a time-domain compensation module to enhance the denoising capability of DADM by compensating for potential losses of global information. Comprehensive experiments demonstrate that DADM significantly outperforms the Novel CNN in denoising EEG signals, showing a notable 7.77% improvement in correlation coefficient (CC) metrics. © 2024, The Authors. All rights reserved.

关键词： Electroencephalography

Pyramid Convolution and Multi-Frequency Spatial Attention for Fine-Grained Visual Categorization

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Xu, Qin Li, Yun Zhang, Mengquan Tao, Zhifu Luo, Bin Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Computer Science and Technology Anhui University Hefei230601 China School of Big Data and Statistics Anhui University Hefei230601 China

Learning diverse detailed feature is crucial for fine-grained visual categorization (FGVC). However, most of existing methods for FGVC use the standard convolution for feature extraction which leads to the loss of many subtle but important features. Besides, in the existing attention models for FGVC, the features are aggregated by a simple global average pooling operation, which is unable to characterize complex feature information. To address these problems, we propose pyramid convolution and multi-frequency spatial attention (PCMFSA) for FGVC. To capture spatial context dependencies of different levels and avoid the loss of the subtle features, pyramid convolution, which uses convolution kernels with various sizes, is introduced in the backbone network. To reduce the influence of background noise and obtain diverse fine-grained feature information, we propose multi-frequency spatial attention (MFSA) module, in which two channel-wise grouping operations are designed and discrete cosine transform (DCT) in the frequency domain is adopted to enhance features of different spatial positions. Moreover, to mine more valuable regions, peak suppression is used to suppress high-response regions after MFSA. A large number of experimental results show that our method achieves state-of-the-art results on four popularly FGVC datasets. © 2022, The Authors. All rights reserved.

关键词： Convolution