The combination of learning analysis technology and terminal equipment can provide a rapidly imitating learning method between devices, and cognitive intelligence affects the learning effect. To enhance automated lear...
详细信息
ISBN:
(纸本)9798350319910
The combination of learning analysis technology and terminal equipment can provide a rapidly imitating learning method between devices, and cognitive intelligence affects the learning effect. To enhance automatedlearning services, we can use federated learning methods to enable the training of different devices and enhance each other through devices. Using the training database to improve the quality of automatedlearning services. In this study, a novel agentassisted active detection anddata collection framework is designed. Monitoring agents can learn from each other to establish intelligent models, and through mutual communication between devices. Can check if established data can be applied to machine data model to get data. It can be used for intelligent manufacturing in the future. The agent may learn methods of learning and managing between devices having different properties. Obtaining experimental simulation and control data, and using machine learning to analyze growth progress and results allow for a deeper analysis of associated adjustments and anticipated changes.
The following paper describes a Deep learning model capable of classifying the inherent readability complexity of a piece of text in European Portuguese. The model was developed using modern Natural Language Processin...
详细信息
ISBN:
(纸本)9783030916077;9783030916084
The following paper describes a Deep learning model capable of classifying the inherent readability complexity of a piece of text in European Portuguese. The model was developed using modern Natural Language Processing techniques, featuring a highly-fine-tuned Neural Network which takes as input both the text as well as multiple metrics relating to it. This classifier was trained on a dataset featuring texts divided in 5 CEFR categories, obtaining an accuracy of 73%, a top 2 accuracy of 90% and an adjacent accuracy of 94%.
Manual classification methods of metamodel repositories require highly trained personnel and the results are usually influenced by the subjectivity of human perception. Therefore, automated metamodel classification is...
详细信息
ISBN:
(纸本)9781728125350
Manual classification methods of metamodel repositories require highly trained personnel and the results are usually influenced by the subjectivity of human perception. Therefore, automated metamodel classification is very desirable and stringent. In this work, Machine learning techniques have been employed for metamodel automated classification. In particular, a tool implementing a feed-forward neural network is introduced to classify metamodels. An experimental evaluation over a dataset of 555 metamodels demonstrates that the technique permits to learn from manually classified data and effectively categorize incoming unlabeled data with a considerably high prediction rate: the best performance comprehends 95.40% as success rate, 0.945 as precision, 0.938 as recall, and 0.942 as F-1 score.
Among the XAI (eXplainable Artificial Intelligence) techniques, local explanations are witnessing increasing interest due to the user need to trust specific black-box decisions. In this work we explore a novel local e...
详细信息
ISBN:
(纸本)9783030916077;9783030916084
Among the XAI (eXplainable Artificial Intelligence) techniques, local explanations are witnessing increasing interest due to the user need to trust specific black-box decisions. In this work we explore a novel local explanation approach appliable to any kind of classifier based on generating masking models. The idea underlying the method is to learn a transformation of the input leading to a novel instance able to confuse the black-box and simultaneously minimizing dissimilarity with the instance to explain. The transformed instance then highlights the parts of the input that need to be (de-)emphasized and acts as an explanation for the local decision. We clarify differences with existing local explanation methods and experiment our approach on different image classification scenarios, pointing out advantages and peculiarities of the proposal.
This paper tackles the classical problem of plagiarism detection by employing current state-of-the-art NLP methods based on Deep learning. We investigate whether transformer models may be used along with existing solu...
详细信息
ISBN:
(纸本)9783030916077;9783030916084
This paper tackles the classical problem of plagiarism detection by employing current state-of-the-art NLP methods based on Deep learning. We investigate whether transformer models may be used along with existing solutions for plagiarism detection, such as Encoplot and clustering, to improve their results. Experimental results show that transformers represent a good solution for capturing the semantics of texts when dealing with plagiarism. Further efficiency improvements are needed as the proposed method is highly effective but also requires high computational resources. This prototype approach paves the way to further fine-tuning such that we may obtain a solution that scales well for a large number of source documents and large-sized documents.
The objective of this study is to classify the states of individuals with bipolar disorder. We employ a dataset that uses the Young Mania Recall Scale to distinguish the manic states of patients as: Mania, Hypo-Mania,...
详细信息
ISBN:
(纸本)9783030916077;9783030916084
The objective of this study is to classify the states of individuals with bipolar disorder. We employ a dataset that uses the Young Mania Recall Scale to distinguish the manic states of patients as: Mania, Hypo-Mania, and Remission. The dataset comprises audio-visual recordings of bipolar disorder patients undergoing a structured interview. Having a small dataset and confidential test labels have motivated us to train a classifier using a semi-supervised ladder network, which benefits from unlabeled data during training. The key advantage of developing a semi-supervised model is removing the manual annotation training data, which is an expensive and time-consuming. We collect informative audio, visual, and textual features from the recordings to realize a multi-model classifier of the manic states. The proposed model achieved a 53.7% UAR and 60.0% UAR on the test and development sets, respectively.
Conventional supervised object recognition methods have been investigated for many years. Despite their successes, there are still two suffering limitations: (1) various information of an object is represented by arti...
详细信息
ISBN:
(纸本)9781479952083
Conventional supervised object recognition methods have been investigated for many years. Despite their successes, there are still two suffering limitations: (1) various information of an object is represented by artificial features only derived from RGB images;(2) lots of manually labeled data is required by supervised learning. To address those limitations, we propose a new semi-supervised learning framework based on RGB and depth (RGB-D) images to improve object recognition. In particular, our framework has two modules: (1) RGB and depth images are represented by convolutional-recursive neural networks to construct high level features, respectively;(2) co-training is exploited to make full use of unlabeled RGB-D instances due to the existing two independent views. Experiments on the standard RGB-D object dataset demonstrate that our method can compete against with other state-of-the-art methods with only 20% labeled data.
This paper presents preliminary results of a research project that aims to build an intelligent Tutoring System (ITS) to learn the tones of the Moore language, the most spoken language in Burkina Faso. This interactiv...
详细信息
ISBN:
(纸本)9781665495196
This paper presents preliminary results of a research project that aims to build an intelligent Tutoring System (ITS) to learn the tones of the Moore language, the most spoken language in Burkina Faso. This interactive learning system is a solution that we propose to meet the needs for local language learning expressed by the staff of government and Non-Governmental Organizations. We use CommonKADS, a knowledge engineering methodology to specify the knowledge and processes of our system, particularly the inference models of the learning activities involved in the domain and pedagogical modules of the ITS.
Gas turbine design is a process that requires designing many interrelated subsystems, e.g., performance, secondary air system, air compression, or combustion. Subsystem models are created by various engineering design...
详细信息
ISBN:
(纸本)9781728151250
Gas turbine design is a process that requires designing many interrelated subsystems, e.g., performance, secondary air system, air compression, or combustion. Subsystem models are created by various engineering design tools. During the design process there exists an extraordinary amount of generated data resulting from created models, simulation, and engine field tests. This data can be leveraged by artificial intelligence techniques such as machine learning to help accelerate the exploration of the large design spaces existing in the complex system of a gas engine. This paper presents a vision and road map of integrating such AIs and preliminary ideas on relevant AI models for such use cases. We explore increasing the realistic nature of existing simulations, approximating simulations to avoid excess computation, and cumulative effect modeling.
Based upon Black Box Variational Inference, a new set of classification algorithms has recently emerged. The goals of this set of algorithms are twofold: 1) increasing generalization power;2) decreasing computational ...
详细信息
ISBN:
(纸本)9783030916077;9783030916084
Based upon Black Box Variational Inference, a new set of classification algorithms has recently emerged. The goals of this set of algorithms are twofold: 1) increasing generalization power;2) decreasing computational and implementation complexity. To this end, we assume a set of latent variables during the generation of data points. We subsequently marginalize the conventional classification likelihood objective function w.r.t this set of latent variables and then apply black-box variational inference to estimate the marginalized likelihood. We evaluate the performance of the proposed method by comparing the results obtained from the application of our method to real-world datasets with those obtained using several classification algorithms. The experimental results prove that our proposed method is competitive.
暂无评论