FAIR data principles represent a set of community-agreed guiding principles and practices for all researchers involved in the eScience ecosystem. the FAIR data principles were created to improve the reuse of data by m...
详细信息
the rapidly growing number of depressed people increases the burden of clinical diagnosis. Due to the abnormal speech signal of depressed patients, automatic audio-based depression recognition has the potential to bec...
详细信息
ISBN:
(纸本)9781450397223
the rapidly growing number of depressed people increases the burden of clinical diagnosis. Due to the abnormal speech signal of depressed patients, automatic audio-based depression recognition has the potential to become a complementary method for diagnosing. However, recognition performance varies largely with different speech acquisition tasks and classifiers, making results not comparable, and the performance requires further improvement before clinical application. this work extracted high-level statistical acoustic features (prosodic, voice-quality, and spectral features) of 23 depressed patients and 29 healthy subjects under spontaneous pronunciation tasks (interview and picture description) and mechanical pronunciation tasks (story reading and word reading), then applied principal component analysis (PCA) to reduce features dimensions, finally employed multilayer perceptron (MLP) to establish the classification model and compared with traditional classifiers (logistic regression, support vector machine, decision tree, and naive Bayes). the results showed that spontaneous pronunciation induced more significantly discriminative acoustic features and achieved better recognition performance accordingly. And the PCA retained 90% useful information with 50% features. Furthermore, MLP achieved the best performance withthe accuracy 0.875 and average F1 score 0.855 under the picture description task. this study provides support for task design and classifier building for audio-based depression recognition, which could assist in mass screening for depression.
Internet of things (IoT) has reshaped our lives by being part of a wide range of fields. thus, using IoT in healthcare became highly demanded. this could be done through monitoring some of the vital health signs that ...
详细信息
Both symbolic and sub-symbolic AI have their limi-tations, but their combination can be more than the sum of their parts. For instance, statistical machine learning has been hugely successful at classification and dec...
详细信息
Many education-related organizations in the U.S., from the National Science Foundation down to local districts, have been pushing to introduce computer science concepts into K-12. Nevertheless, many students complete ...
详细信息
Partial code usually involves non-fully-qualified type names (non-FQNs) and undeclared receiving objects. Resolving the FQNs of these non-FQN types and undeclared receiving objects (referred to as type inference) is t...
ISBN:
(纸本)9781450394758
Partial code usually involves non-fully-qualified type names (non-FQNs) and undeclared receiving objects. Resolving the FQNs of these non-FQN types and undeclared receiving objects (referred to as type inference) is the prerequisite to effective search and reuse of partial code. Existing dictionary-lookup based methods build a symbolic knowledge base of API names and code contexts, which involve significant compilation overhead and are sensitive to unseen API names and code context variations. In this paper, we formulate type inference as a cloze-style fill-in-blank language task. Built on source code naturalness, our approach fine-tunes a code masked language model (MLM) as a neural knowledge base of code elements with a novel “pre-train, prompt and predict” paradigm from raw source code. Our approach is lightweight and has minimum requirements on code compilation. Unlike existing symbolic name and context matching for type inference, our prompt-tuned code MLM packs FQN syntax and usage in its parameters and supports fuzzy neural type inference. We systematically evaluate our approach on a large amount of source code from Github and Stack Overflow. Our results confirm the effectiveness of our approach design and the practicality for partial code type inference. As the first of its kind, our neural type inference method opens the door to many innovative ways of using partial code.
the PBFT algorithm is a consensus algorithm that reduces communication complexity to a polynomial level, addressing the inefficiency of traditional Byzantine fault tolerance mechanisms. It is commonly adopted in the c...
详细信息
ISBN:
(纸本)9798400708909
the PBFT algorithm is a consensus algorithm that reduces communication complexity to a polynomial level, addressing the inefficiency of traditional Byzantine fault tolerance mechanisms. It is commonly adopted in the consensus layer of blockchain systems. However, as the number of system nodes increases, the algorithm faces the challenge of rapidly increasing communication complexity. One solution to this problem is to use deterministic algorithms to select smaller consensus groups. However, in an open network, this approach is susceptible to Distributed Denial of Service (DDoS) attacks against the main nodes, posing a risk of system unavailability. Another solution involves using Verifiable Random Functions (VRF) technology to select consensus group nodes. However, this approach often introduces additional time overhead. To combine the advantages of both solutions, this paper proposes an improved PBFT algorithm based on a Dual-Random-Selection mechanism (DRS-BFT). the algorithm uses an unbiased random selection function and a verifiable random function to select consensus group members and the primary node. In situations with a large number of nodes, the algorithm enhances consensus efficiency by selecting a consensus group, minimizing additional communication overhead, and ensuring system security. Security analysis demonstrates that the system, under the assumption of security requirements, can guarantee data integrity, consistency, and availability. It can also prevent targeted attacks against the primary node that could lead to system unavailability. To validate the system's functionality and performance, the paper designs and develops a prototype system and establishes an experimental environment. the experiments show that the algorithm dynamically adjusts the consensus group range through the Dual-Random-Selection mechanism to adapt to changes in network conditions. Additionally, compared to traditional PBFT, the algorithm exhibits superior performance.
In this work, the authors present the study from personal experience to gain some insight into the different aspects of efficient self-directed learning. Two case studies regarding the self-directed learning approach ...
详细信息
software is playing an increasingly important role in modern radar software. software faults often lead to serious system failure. A fault prevention approach is proposed for radar software. the proposed approach foll...
详细信息
Over the past decades, the development of predictive maintenance strategies, like Prognostics and Health Management (PHM), have brought new opportunities to the maintenance domain. However, implementing such systems a...
详细信息
ISBN:
(纸本)9789897584008
Over the past decades, the development of predictive maintenance strategies, like Prognostics and Health Management (PHM), have brought new opportunities to the maintenance domain. However, implementing such systems addresses several challenges. First, all information related to the system description and failure definition must be collected and processed. In this regard, using an expert system (ES) seems interesting. the second challenge, when monitoring complex systems, is to deal withthe high volume and velocity of the input data. To reduce them, Complex Event Processing (CEP) can be used to identify relevant events, based on predefined rules. these rules can be extracted from the ES knowledge base using model transformation. this process consists in transforming some concepts from a source to a target model using transformation rules. In this paper, we propose to transform a part of the knowledge from a condition-based maintenance (CBM) model into CEP rules. After further explaining the motivations behind this work and defining the principles behind model-driven architecture and model transformation, the transformation from a CBM model to a "generic rules" model will be proposed. this model will then be transformed into an Event Processing Language (EPL) model. Examples will be given as illustrations for each transformation.
暂无评论