Search

Data augmentation strategies have always been important in machine learning techniques and play a unique role in model performance optimization processes. Therefore, in recent years, these techniques have become popular in the artificial intelligence field. In this paper, a new data augmentation strategy is proposed based on the interpretation algorithm of deep convolutional neural networks, i.e., constructing new training samples by deeply exploiting key features extracted from interpretable networks to achieve sample augmentation. Thus, a novel supervised data augmentation approach known as Supervised Data Augmentation–Key Feature Extraction (SDA-KFE) was proposed. By introducing the Neural Network Interpreter-Segmentation Recognition and Interpretation (NNI-SRI) algorithm, an augmentation strategy is proposed that can balance the high accuracy and high robustness of the final model while ensuring a large amount of data augmentation. The advantages of the SDA-KFE algorithm are mainly reflected in the following aspects. First, it is easy to implement. This algorithm is implemented based on the lightweight NNI-SRI algorithm, which lays the foundation for the implementation of SDA-KFE so that it can be easily implemented on convolutional neural networks. Second, this model, which is widely applicable, can be applied to almost any deep convolutional network. Through research and experiments on this proposed algorithm, SDA-KFE can be applied in graphical image binary classification and multiclassification models. Third, SDA-KFE can rapidly construct data samples with diverse variations. Under the premise of determining the classification labels of the generated samples, the distribution of the feature unit composition of the samples can be controlled. Compared with traditional data augmentation methods, SDA-KFE can control the direction of the model performance, i.e., the balance between the pursuit of high accuracy and robust performance of the model. Therefore, the novel supervised augmentation approach proposed in this paper is relevant for optimizing deep convolutional neural networks, solving model overfitting, augmenting data types, etc. The data augmentation algorithm proposed in this paper can be regarded as a useful supplement to traditional data augmentation methods, such as horizontal or vertical image flipping, cropping, color transformation, extension and rotation. [ABSTRACT FROM AUTHOR]

Perceptron Neural Networks (PNNs) are essential components of intelligent systems because they produce efficient solutions to problems of overwhelming complexity for conventional computing methods. Many papers show that PNNs can approximate a wide variety of functions, but comparatively, very few discuss their limitations and the scope of this paper. To this aim, we define two classes of Boolean functions – sensitive and robust –, and prove that an exponentially large set of sensitive functions are exponentially difficult to compute by multi-layer PNNs (hence incomputable by single-layer PNNs). A comparatively large set of functions in the second one, but not all, are computable by single-layer PNNs. Finally, we used polynomial threshold PNNs to compute all Boolean functions with quantum annealing and present in detail a QUBO computation on the D-Wave Advantage. These results confirm that the successes of PNNs, or lack of them, are in part determined by properties of the learned data sets and suggest that sensitive functions may not be (efficiently) computed by PNNs. [ABSTRACT FROM AUTHOR]

Multimodal emotion recognition, that is, emotion recognition uses machine learning to generate multi-modal features on the basis of videos which has become a research hotspot in the field of artificial intelligence. Traditional multi-modal emotion recognition method only simply connects multiple modalities, and the interactive utilization rate of modal information is low, and it cannot reflect the real emotion under the conflict of modal features well. This article first proves that effective weighting can improve the discrimination between modalities. Therefore, this paper takes into account the importance differences between multiple modalities, and assigns weights to them through the importance attention network. At the same time, considering that there is a certain complementary relationship between the modalities, this paper constructs an attention network with complementary modalities. Finally, the reconstructed features are fused to obtain a multi-modal feature with good interaction. The method proposed in this paper is compared with traditional methods in public datasets. The test results show that our method is accurate in It performs well in both the rate and confusion matrix metrics. [ABSTRACT FROM AUTHOR]

• Many ML algorithms are used to identify the main determinants of a bank default. • We use of a graph neural network that has never been used in a financial context. • We obtain a balanced dataset by customizing the heuristic oversampling method. • Like previous literature, we show that neural network outperforms other approaches. • We include, for the first time, competition among the possible default determinants. The financial sector is very interested in Artificial Intelligence due to the opportunities that it offers, especially those related to methods of machine-learning. The aim of this paper is to employ a variety of machine-learning algorithms to identify the main determinants of bank default and to understand the impact of each variable on it. Bank default is one of the most studied topics in financial literature because of the severity of its consequences on the whole economic system. However, little attention has been paid to the identification of the major determinants of bank failures via machine-learning approaches. This paper employs several machine-learning algorithms, including a graph neural network that has never been used in a financial context. Another novelty is the implementation of a balanced dataset by customising the heuristic oversampling method based on k-means and synthetic minority over-sampling technique. This paper also deals with the inclusion of competition among the possible default determinants. The dataset consists of all the banks in the Euro Area in the period 2018–2020. The results obtained are useful from both micro- and macro-economic points of view. [ABSTRACT FROM AUTHOR]

One of the negative aspects of the world becoming more digitized has been fake news, i.e., online disinformation – false, often fabricated reports of events, written and read on websites. The term has already entered collective consciousness and become an inseparable element of scientific discourse. Once a piece of news goes online, stopping it from spreading may become a complicated matter. Literature suggests that the two main pillars of the effective fight against fake news are education and detection. Thus, this paper describes a multidisciplinary study performed by a group of scientists representing two distinct fields - AI and linguistics. In their joint study, they compared, formally evaluated and explored the intersection between two approaches to fake news detection, i.e., the automated one, using a machine-learning-based tool, and the linguistic-based human judgement, using the data from two disinformation campaigns, sourced from two open benchmark fake news datasets. The study focused on the news' headlines as an effective proxy for the identification of fake news. In accordance with the achieved results, the paper argues that in the fight against fake news, the two approaches have the potential of augmenting and enhancing each other, utilizing the state-of-the-art technologies and linguistic knowledge. In addition, this paper provides a list of the linguistic features characteristic of possible disinformation, which is the most comprehensive collection of this kind in the subject literature to date. [ABSTRACT FROM AUTHOR]

This paper introduces the Ulimisana Optimisation Algorithm enabled Population Based Training (PBT-UOA) framework which allows for hyperparameters to be fine-tuned using a population based meta-heuristic algorithm at the same time as parameters are being optimised. Models are trained until near-convergence on the updated hyperparameters and the parameters of the best performing model are shared to warm start the other models in the next hyperparameter tuning iteration. In the PBT-UOA, all models are trained using the same dataset. This framework performed better than the Bayesian Optimisation algorithm. This paper also introduces the Ulimisana Optimisation Algorithm enabled Federated Learning (FL-UOA) framework which is an extension of the PBT-UOA. This framework is introduced to address the challenges of scattered datasets and privacy that is presented by the increase in connected end-devices. The FL-UOA learns on local data in scattered end-devices without sending datasets to a central server. The training datasets in local end-devices are used to evaluate models trained in other end-devices. The performance metrics are used to update the Social Trust Network (STN) of the FL-UOA framework. The FL-UOA outperformed the classic Federated Learning framework. This STN updating technique was tested in Machine Learning (ML) Unfairness to see how well it functioned as a regularisation term. This was achieved by training different models on subsets that contained datasets representing only specific sensitive groups. Results showed that by updating the hyperparameters while learning the parameters on the dataset scattered across different devices, the FL-UOA, takes advantage of diversified learning and reduces the ML Unfairness for models trained on group specific datasets. [ABSTRACT FROM AUTHOR]

• Fuzzy knowledge graph is a more general description of classical knowledge graph. • The fuzzy semantic scalability between multiple associative predicates is analyzed. • The mathematical model of semantic relationship in fuzzy knowledge graph is designed. • Some fuzzy reasoning rules are presented to realize fuzzy semantic extension. • Performance of the strategy discovers more implicit valid knowledge with fuzzy semantic. As the latest achievement of the development in semiotics, knowledge graph has been recognized and widely used by more and more researchers for its rich semantic information and clear logical structure. How to discovery the deep relevant knowledge from the massive graph-structured data has become a hot spot of artificial intelligence. Considering that some predicates in knowledge graph express fuzzy relationships whose semantics are not certain, the basic schema of classical knowledge graph in the form of RDF triple cannot describe the fuzzy semantic information effectively. To counter above problems, in this paper, we present a new semantic representation and reasoning model for multiple associative predicates by introducing fuzzy theory. Concretely, the presented method defines a new fuzzy annotating strategy to represent the fuzzy semantics between associative predicates in different RDF triples. On this basis, some fuzzy reasoning rules are presented to realize fuzzy semantic extension for classical knowledge graph. Lastly, the experimental results show that our proposal can discover more implicit valid knowledge with fuzzy semantic and have a good consistency with the intuition of human judgments. Overall, the methods proposed in this paper constitute some effective ways of knowledge discovery of structured semantic data. [ABSTRACT FROM AUTHOR]

Simplicial map neural networks (SMNNs) are topology-based neural networks with interesting properties such as universal approximation ability and robustness to adversarial examples under appropriate conditions. However, SMNNs present some bottlenecks for their possible application in high-dimensional datasets. First, SMNNs have precomputed fixed weight and no SMNN training process has been defined so far, so they lack generalization ability. Second, SMNNs require the construction of a convex polytope surrounding the input dataset. In this paper, we overcome these issues by proposing an SMNN training procedure based on a support subset of the given dataset and replacing the construction of the convex polytope by a method based on projections to a hypersphere. In addition, the explainability capacity of SMNNs and effective implementation are also newly introduced in this paper. [ABSTRACT FROM AUTHOR]

• This paper proposes a human–machine integration ECG intelligent annotation system. • The ECG intelligent annotation system can free the annotation experts from the heavy manual annotation work. • This paper proposes an accurate beat generation model suitable for all beat types. • The generated beat data can be used as a supplement to labelled data. • This paper proposes an intelligent beat pre-annotation model based on intelligent simulation generated data. • The intelligent beat pre-annotation model achieved the best performance. An electrocardiogram (ECG) consists of complex segments, such as P-QRS-T waves. Manual ECG annotation is challenging and time-consuming, even for specialist physicians. The shortage of labelled ECG data is one of the essential factors that affect ECG intelligent analysis's long-term development. This study proposes an intelligent ECG-assisted annotation system, that not only supplements labelled data, but also significantly reduces the workload compared with manual annotation. Since beat annotation is the most basic and important part, a GAN-based generation model that can generate 14 types of simulation beats and a CNN-based beat pre-annotation model are proposed. The experimental results show that the simulation beat has high similarity to real beat and the accuracy of the pre-annotation model on the test set of 14 classes of beats is 99.28%. The proposed ECG intelligent annotation system's self-learning mechanism could improve pre-annotation performance and annotation efficiency by generating more labelled data. The proposed annotation system can also be extended to other data annotation applications. [ABSTRACT FROM AUTHOR]

• An AI algorithm for data enhanced encryption at the end and the intermediate nodes of autonomous IoT systems is proposed. • An AI access strategy is designed for reducing the calculation cost of data encryption. • A data shared matrix is proposed to share the encrypted data to achieve the (k, n) threshold strategy. Aiming at the security issues during the multi-types data storage and data transmission in autonomous Internet of Things (IoT) systems, this paper proposes an AI algorithm for data enhanced encryption used in the ends and the intermediate nodes of IoTs. The algorithm in this paper first constructs a three-dimensional Arnold transformation matrix for data unit value encryption in the end of IoTs, and designs a quantum logic intelligent mapping that effectively diffuses the encrypted data units to reduce the linear correlation of the image data and to improve the security performance of IoT edge data. Furthermore, the algorithm designs an AI access strategy for scrambling sequence nodes and builds a random-access route for the elements of the scrambling sequence which can reduce the calculation cost and improve the operating efficiency of IoT system in the ends and intermediate nodes. Finally, the data shared matrix is used to share the encrypted data to achieve the (k, n) threshold strategy. Experimental results prove that the algorithm has high plaintext and key sensitivity and can effectively resist brute force attacks, statistical analysis and differential attacks. The algorithm in this paper provides an AI solution for data security encryption in the ends and the intermediate nodes of autonomous IoT systems. [ABSTRACT FROM AUTHOR]

In a previous paper (Kim, 2019), an analytical framework based on the constraint satisfaction problems was proposed to reveal the characteristics of households using event logs from smart door lock systems. This work provides a more rigorous justification for the previous approach. This paper proposes a novel parameter estimation method called the maximum feasibility estimation (MFE). The MFE does not rely on any assumption about the parametric family of probability densities from which a random observation is drawn. Instead, we assume that constraints are imposed on observations and that some of the constraints are a function of a parameter of interest. The proposed estimator maximizes the feasible region, a set of all possible observations that satisfy those constraints. The method proposed is validated using synthetic data as well as real streaming event log data. [ABSTRACT FROM AUTHOR]

To address the challenge of facilitating overall problem solving and improving the overall efficiency of intelligent systems efforts through group decision making (GDM) by experts at all stages of the systems, this paper proposes an approach that focuses on improving consistency and enhancing local consensus. The proposed approach specifically considers the probabilistic linguistic preference relations (PLPRs) and incorporates personalized individual semantics (PISs) of decision makers (DMs). For the consistency procedure, we construct an expectation-additive consistency-driven semantic model to acquire PISs of different DMs. Secondly, the preference relation (PR) which does not satisfy the consistency is improved based on a minimum adjustment model. In particular, according to the expected value with PISs this paper, DMs offer PLPRs can be transformed into fuzzy preference relations (FPRs) correspondently for making decisions. For the FPRs, a virtue consensus measure is explored for the consensus reaching process (CRP) from two levels to combine the average value and variance of pair similarities, which effectively improves the accuracy of the consensus measurement. Next, a collective consensus level is calculated to replace the consensus threshold objectively, and then the DMs and alternative pairs that do not reach consensus are identified locally from two levels. Subsequently, an optimisation model that combines the two objectives is developed to improve the consensus level while ensuring consistency. Finally, our method is applied to a publicly available air quality dataset, and the consensus and ranking results are discussed in comparison with other advanced methods to confirm the superiority of the established method. [ABSTRACT FROM AUTHOR]

Location factor of recommender systems has been extensively studied in the past decade. However, there is no research thoroughly analyzing location's role in news recommendation. In this paper, a comprehensive exploration on role of location in news recommendation is presented. First of all, based on analysis of real news datasets, we find that news recommendation differs from spatial item recommendation. Location affects news consumption behaviors of users with two-fold aspects including geographic feature and semantic feature. Regarding geographic feature, location influences news recommendation according to region rather than latitude-longitude level. Furthermore, interesting news topics are also impacted by semantic feature of location. Semantic feature may play a more positive role than geographic feature. The novel findings consistently manifest that, as non-spatial items, news differ from spatial items in that location influences users' selection in terms of different pattern and degree. In summary, geographic and semantic features influence reading preference through mapping locations into special topics. Changing of location topics leads to varying of reading preference. The news datasets in this paper belong to check in data. NewsREEL dataset is from a company, and it is provided by German researcher. The location data in Twitter dataset is also check in data. NetEase news dataset are collected from NetEase news websites, and the type of location data is city or region. [ABSTRACT FROM AUTHOR]

How to derive an investor's preference is vital for investment advisors and online lending platforms for targeted marketing strategies, e.g., market segmentation and financial product recommendation. However, investor preference analysis usually depends on judgments from human investment experts, which are inherently subjective and costly. Intelligent investment advisors (or Robo-advisors), supported by cutting-edge technologies such as machine learning and artificial intelligence, are established to relieve these pending issues. This paper employs an online optimization framework to obtain investors' preferences for further financial product recommendations. This proposed method allows us to update the investor's preference for newly-arriving data sets and tackle the situation where plenty of missing values in investors' records are present. Unlike the black-box-like machine learning approach, our method can provide more managerial implications regarding why one financial product/service is preferred. Real-world data set from an online financial platform is used to compare the existing approaches and shows the stronger and more stable performance of our method when facing different data-missing types and situations with different missing degrees, followed by a recommendation system application. [ABSTRACT FROM AUTHOR]

Explainable artificial intelligence (XAI) is gradually becoming a key component of many artificial intelligence systems. However, such pursuit of transparency may bring potential privacy threats to the model confidentially, as the adversary may obtain more critical information about the model. In this paper, we systematically study how model decision explanations impact model extraction attacks, which aim at stealing the functionalities of a black-box model. Based on the threat models we formulated, an XAI-aware model extraction attack (XaMEA), a novel attack framework that exploits spatial knowledge from decision explanations is proposed. XaMEA is designed to be model-agnostic: it achieves considerable extraction fidelity on arbitrary machine learning (ML) models. Moreover, we proved that this attack is inexorable, even if the target model does not proactively provide model explanations. Various empirical results have also verified the effectiveness of XaMEA and disclosed privacy leakages caused by decision explanations. We hope this work would highlight the need for techniques that better trade off the transparency and privacy of ML models. • We propose XaMEA, three XAI-aware model extraction attack architectures. • We further carry out XAI-aware model extraction attacks against non-explanation target models. • We evaluate the attack effectiveness of XaMEA with an exhaustive set of experiments. [ABSTRACT FROM AUTHOR]

This paper presents a method called Interval Incremental Learning (IIL) to capture spatial and temporal patterns in uncertain data streams. The patterns are represented by information granules and a granular rule base with the purpose of developing explainable human-centered computational models of virtual and physical systems. Fundamentally, interval data are either included into wider and more meaningful information granules recursively, or used for structural adaptation of the rule base. An Uncertainty-Weighted Recursive-Least-Squares (UW-RLS) method is proposed to update affine local functions associated with the rules. Online recursive procedures that build interval-based models from scratch and guarantee balanced information granularity are described. The procedures assure stable and understandable rule-based modeling. In general, the model can play the role of a predictor, a controller, or a classifier, with online sample-per-sample structural adaptation and parameter estimation done concurrently. The IIL method is aligned with issues and needs of the Internet of Things, Big Data processing, and eXplainable Artificial Intelligence. An application example concerning real-time land-vehicle localization and tracking in an uncertain environment illustrates the usefulness of the method. We also provide the Driving Through Manhattan interval dataset to foster future investigation. [ABSTRACT FROM AUTHOR]

• Providing AI-enabled IoT system in healthcare monitoring and control. • Adjusting the depth of anesthesia in surgery by automatically infusion. • Designing an adaptive control system using a robust control method and fuzzy system. • Employing fuzzy learning to provide an intelligent estimator for patient model uncertainties. • Remote tuning of drug infusion through network channels. Smart remote patient monitoring and early disease diagnosis systems have made huge progresses after the introduction of Internet of Things (IoT) and Artificial Intelligence (AI) concepts. This paper proposes an AI-enabled IoT system to monitor and adjust the depth of anesthesia via network channels. More precisely, fuzzy learning systems are employed to develop a control system for the depth of anesthesia in surgeries. This scheme is composed of variable structure control and adaptive type-II fuzzy systems. Therefore, the controller is adaptive and robust to any perturbations and disturbances that may happen during a patient's surgery. The adaptive type-II fuzzy system is designed as an intelligent online estimator to approximate patient model uncertainties. This estimation helps in boosting the performance of the variable structure control system. An artificial neuron is also designed to reduce chattering for the proposed control system. The designed control system can efficiently adjust the anesthesia drug infusion rate and regulate the Bispectral index. The networked structure of the proposed system makes remote tuning of drug infusion possible. Performance of the designed controller is evaluated on several patient models. Simulation results confirm the validity and effectiveness of the proposed remote drug delivery system. [ABSTRACT FROM AUTHOR]

Many applications in cognitive science and artificial intelligence utilize semantic similarity and relatedness to solve difficult tasks such as information retrieval, word sense disambiguation, and text classification. Previously, several approaches for evaluating concept similarity and relatedness based on WordNet or Wikipedia have been proposed. WordNet-based methods rely on highly precise knowledge but have limited lexical coverage. In contrast, Wikipedia-based models achieve more coverage but sacrifice knowledge quality. Therefore, in this paper, we focus on developing a comprehensive semantic similarity and relatedness method based on WordNet and Wikipedia. To improve the accuracy of existing measures, we combine various taxonomic and non-taxonomic features of WordNet, including gloss, lemmas, examples, sister-terms, derivations, holonyms/meronyms, and hypernyms/hyponyms, with Wikipedia gloss and hyperlinks, to describe concepts. We present a novel technique for extracting ' is-a ' and ' part-whole ' relationships between concepts using the Wikipedia link structure. The suggested technique identifies taxonomic and non-taxonomic relationships between concepts and offers dense vector representations of concepts. To fully exploit WordNet and Wikipedia's semantic attributes, the proposed method integrates their semantic knowledge at feature-level, combining semantic similarity and relatedness into a single comprehensive measure. The experimental results demonstrate the effectiveness of the proposed method over state-of-the-art measures on various gold standard benchmarks. [ABSTRACT FROM AUTHOR]

Acute hypotensive episode (AHE) is a common serious postoperative complication in ICU, which may raise multiple system failure (especially of cardiac and respiratory kinds), and even cause death. Timely and effective clinical intervention is obviously vital to the saving of patients. AHE detection involves physiological time-series monitoring, processing and prediction technologies, which can offer insights to neuroscientists, biologists, and even provide support for clinicians. This paper presents a hybrid artificial intelligence model combined with CEEMDAN (complete ensemble empirical mode decomposition with adaptive noise, a typical method for physiological signal decomposition), deep learning, multiple gene expression programming and fuzzy expert system for AHE detection. In this paper, the physiological signal is selected from a benchmark dataset, for example MIMIC-II (Multiparameter Intelligent Monitoring in Intensive Care II), which collects large scale real patients' data for clinical research. In the hybrid model, a typical signal decomposition method is employed for AHE signal processing, and an autoencoder based deep neural network is established for feature extraction. Finally, a reliable and explainable classifier is presented by fusing gene expression programming and the fuzzy method. Experimental results based on real data set demonstrate that the proposed method outperforms state-of-the-art AHE detection methods by achieving the prediction accuracy of 88.14% in 2866 records. [ABSTRACT FROM AUTHOR]

• Explainable framework for meta-learning. • Efficiency and high causality. • Intervention and counterfactual. With the growing convergence of artificial intelligence and daily life scenarios, the application scenarios for intelligent decision methods are becoming increasingly complex. The development of various machine learning algorithms has benefited all disciplines of study, but choosing which algorithm is most suitable for a certain problem among a large number of algorithms is a challenge that every field must overcome. Another challenge at the practical application level is that machine learning algorithms currently trained with large amounts of data are primarily black-box and uninterpretable. This indicates that these methods pose potential risks and are difficult to rely on, thus hindering their application in sensitive fields such as finance and healthcare. The first challenge can be overcome by using meta-learning to combine data and prior knowledge to efficiently and automatically select the machine learning models. The second challenge remains to be addressed due to the lack of interpretability of traditional meta-learning techniques and deficiencies in transparency and fairness. Achieving the interpretability of meta-learning in autonomous algorithm selection for classification is crucial to balance the need for high accuracy and transparency of machine learning models in practical application scenarios. This paper proposes EFFECT , an interpretable meta-learning framework that can explain the recommendation results of meta-learning algorithm selection and provide a more complete and accurate explanation of the recommendation algorithm's performance on specific datasets combined with business scenarios. Extensive experiments have demonstrated the validity and correctness of this framework. [ABSTRACT FROM AUTHOR]

• Explanation Sets, a new framework that unifies counterfactuals and semifactuals. • Counterfactuals and semifactuals are defined in terms of a similarity measure. • Restrictions and preferences over the explanations are defined using a neighborhood. • A taxonomy for set-based representations. Explainable Machine Learning (ML) is an emerging field of Artificial Intelligence that has gained popularity in the last decade. It focuses on explaining ML models and their predictions, enabling people to understand the rationale behind them. Counterfactuals and semifactuals are two instances of Explainable ML techniques that explain model predictions using other observations. These techniques are based on the comparison between the observation to be explained and another one. In counterfactuals, their prediction is different, and in semifactuals, it is the same. Both techniques have been studied in the Social Sciences and Explainable ML communities, and they have different use cases and properties. In this paper, the Explanation Set framework, an approach that unifies counterfactuals and semifactuals, is introduced. Explanation Sets are example-based explanations defined in a neighborhood where most observations satisfy a grouping measure. The neighborhood allows defining and combining restrictions. The grouping measure determines if the explanations are counterfactuals (dissimilarity) or semifactuals (similarity). Besides providing a unified framework, the major strength of the proposal is to extend these explanations to other tasks such as regression by using an appropriate grouping measure. The proposal is validated in a regression and classification task using several neighborhoods and grouping measures. [ABSTRACT FROM AUTHOR]

As the basis of many artificial intelligence tasks, text similarity measurement has received extensive attention in current studies. However, few of them focus on the resource-limited scenes (i.e., limited computational resources and few training datasets), which are becoming increasingly popular and challenging with the development of the Internet of Things. Worse still, popular methods such as the deep-neural-network-based methods may lose their power in such scenes, since they typically require considerable computational resources. As for most current traditional methods, they also have issues of not effectively exploiting the semantic information in the sentences. As an alternative, this paper proposes a lightweight and semantically rich text similarity measurement model named the TES-TK model. In this model, a sentence is first transformed into a tree structure called TES-Tree with the integration of syntactic information, semantic knowledge, and topic distribution, aiming to comprehensively represent the multidimensional semantics of sentences. Afterward, a modified tree kernel model is designed to calculate the similarity between each pair of TES-Trees. In this way, the similarity score between the two related sentences can be retrieved. Experiments on 19 public benchmark datasets (STS2012–2015) demonstrate that the proposed approach exhibits significantly better performance than the compared eight peer methods on most datasets. Especially in resource-limited scenes, our approach achieved highly competitive results compared with the latest methods, such as BERT. [ABSTRACT FROM AUTHOR]

Fuzzy inference with the Bandler-Kohout subproduct (BKS) has been successfully applied in many fields such as fuzzy control, artificial intelligence, image processing, data mining, decision-making, prediction, classification and so on. However, one has to face with the rule explosion in these applications. To deal with this problem, hierarchical fuzzy systems with the compositional rule of inference (CRI) method have been constructed by a series of low-dimensional sub fuzzy systems. And it has been proved that hierarchical fuzzy inference method can efficiently restrain the explosion of fuzzy rules. Therefore, in order to increase the computational efficiency of the fuzzy inference based on the BKS when multi-input-single-output (MISO) fuzzy rules are involved, this paper mainly constructs two hierarchical fuzzy inference methods based on the BKS in which the if-then rules are respectively interpreted by fuzzy implications and ML -implications. Moreover, the validity of the two BKS hierarchical fuzzy inferences is studied with the GMP rules. Finally, two examples are employed to illustrate the computational efficiency of our proposed BKS hierarchical inference methods. [ABSTRACT FROM AUTHOR]

In response to fighting COVID-19 pandemic, researchers in machine learning and artificial intelligence have constructed some medical knowledge graphs (KG) based on existing COVID-19 datasets, however, these KGs contain a considerable amount of semantic relations which are incomplete or missing. In this paper, we focus on the task of knowledge graph embedding (KGE), which serves an important solution to infer the missing relations. In the past, there have been a collection of knowledge graph embedding models with different scoring functions to learn entity and relation embeddings published. However, these models share the same problems of rarely taking important features of KG like attribute features, other than relation triples, into account, while dealing with the heterogeneous, complex and incomplete COVID-19 medical data. To address the above issue, we propose a graph feature collection network (GFCNet) for COVID-19 KGE task, which considers both neighbor and attribute features in KGs. The extensive experiments conducted on the COVID-19 drug KG dataset show promising results and prove the effectiveness and efficiency of our proposed model. In addition, we also explain the future directions of deepening the study on COVID-19 KGE task. [ABSTRACT FROM AUTHOR]

Information modeling and handling in uncertain environments is an important topic in the field of modern artificial intelligence. In practical applications of classification problems, the data harvested by the agent is usually not precise. Based on multi-valued mapping of probabilities expressed by Basic Probability Assignment (BPA), Dempster-Shafer Theory (DST) has a strong ability to model and handle uncertain information. In this paper, we propose a method of fusing attributes to enhance the quality of uncertain data under the framework of DST. The fusion method is based on proposed uncertainty and dissimilarity measures, which performs consistent transformations on belief information in DST. We simulate uncertain data by adding different noises to precise datasets and classify the improved data using common classifiers. With the increasing uncertainty degree of data, the proposed method has higher accuracy and robustness than other methods. [ABSTRACT FROM AUTHOR]

Analysing the interpretability of a fuzzy system (either hierarchical or not) involves consideration of its semantic properties and evaluation of its structural complexity. The present paper concentrates on the semantic aspects of interpretability in hierarchical fuzzy systems. Complexity reduction is also considered, but from the perspective of its interaction with semantic preservation. In that sense, the paper shows that only the use of intermediate variables with meaning (interpretable variables) will transform the complexity reduction into a real improvement in interpretability. The paper formalises this idea by introducing the concept of semantically decouplable hierarchies. Under the assumption of semantic decouplability, it is shown that the interpretability of the overall hierarchical system can be directly obtained from that of its subsystems. Consequently, the paper defines and measures the interpretability of a semantically decouplable hierarchical fuzzy system as the aggregation of the interpretability of its subsystems. Finally, several options will be considered for this aggregation process. [ABSTRACT FROM AUTHOR]

Highlights • This paper firstly propose NMI-sub-models selection-based model. • This paper firstly used artificial intelligence algorithms and TA to search the optimal size of the neighborhood in the neighborhood mutual information method. • Three cases are used to validate the proposed model. • The proposed NMI-MRMR-CS algorithm is better than the other approaches. Abstract Combination models are regarded as a popular approach to improve forecasting accuracy. Determining how to select optimal sub-models from all possible candidate models is considered to be a challenge in combination models because of the high computational overhead. However, there is fewer studies on sub-models selection for combination approaches. To this end, this study proposes a novel sub-models selection algorithm named neighborhood mutual information (NMI) theory-based maximizing linear relevance and minimizing linear redundancy (NMI-MRMR) to select the optimal sub-models set. Cuckoo search (CS) is used to find the optimal parameter of a neighborhood for the proposed algorithm by avoiding the evaluation of all possible parameters. Three cases are used to validate the performance of the proposed approach. The experimental forecasting results demonstrate that the combination of the individual models selected by the NMI-MRMR based CS algorithm significantly outperforms the combination of the individual models selected by traditional approaches. Furthermore, a comparison with two other artificial intelligence algorithms and a traversal algorithm indicate that the CS algorithm is more suitable for choosing the optimal parameter for the proposed algorithm. The proposed approach not only overcomes the shortcomings of combining all individual models but is also able to select the optimal sub-models set. [ABSTRACT FROM AUTHOR]

Technology has grown rapidly in recent years, and new solutions that rely on Machine Learning (ML) and Artificial Intelligence (AI) are introduced every day. With such fast-paced advancement, inspecting and fully comprehending how given models make decisions is becoming problematic. The complex decision-making process of these models has become a black box, making it challenging to unravel how they work; therefore, eXplainable Artificial Intelligence (XAI) methods are crucial for further development. This paper discusses how state-of-the-art techniques determine classifications and why they need to be revised to understand the prediction-generating process fully. It compares those existing solutions with the new method called Principal Image Sections Mapping - PRISM, which relies on Principal Component Analysis and allows visualising the most significant features recognised by a given Convolutional Neural Network. PRISM is implemented in a piece of software called TorchPRISM that can generate and present the clustering based on the method's output. The result can indicate ambiguous classes discrimination; thus, the possibility of automating the output analysis process is also discussed. The paper's main objective is to examine how PRISM enhances the current understanding of the decision-making process and introduce a tool that can facilitate analysing the output. PRISM implementation (TorchPRISM) can be found in the public GitHub repository: https://github.com/szandala/TorchPRISM • PRISM method can explain CNN's decisions in an understandable way for various users. • PRISM uses PCA to point features and highlights them using an RGB-coloured image mask. • PRISM's effectiveness is assessed based on qualitative and quantitative measures. • TorchPRISM software allows the identification of ambiguous classes in large datasets. • PRISM overcomes the deficiencies of existing eXplainable AI and saliency map methods. [ABSTRACT FROM AUTHOR]

We develop an agent-based artificial general intelligent system that can be implemented in compact and power-efficient electronic hardware. The hardware under development is called the Street Engine, which is a hardware-based cognitive architecture for implementing agent-based artificial intelligence. In this paper, we introduce an agent-based system to replicate simple cognitive behaviours. In the processes of this system, numerical data are converted into fuzzy symbolic representations of the surrounding environment, and reasoning rules are included in a modified Fuzzy Inference System to support the cognitive decision-making. We use a case study example, the homing behaviour of the honey bee, to demonstrate constructing production rules and implementing the cognitive and reasoning capabilities of agents. The low level cognitive behaviour is converted into a rule-based fuzzy system, and hardware-based experiments have been conducted to verify the effectiveness of the proposed technique. [ABSTRACT FROM AUTHOR]

Knowledge tracing (KT) has become an increasingly relevant problem in intelligent education services, which estimates and traces the degree of learner's mastery of concepts based on students' responses to learning resources. The existing mainstream KT models, only attribute learners' feedback to the degree of knowledge mastery and leave the influence of mental ability factors out of consideration. Although ability is an essential component of the problem-solving process, these knowledge-centered models cause a contradiction between data fitting and rationalization of the model decision-making process, making it difficult to achieve high precision and readability simultaneously. In this paper, an innovative KT model, a bility b oosted k nowledge t racing (ABKT) 1 1 Our implementations are available in https://github.com/ccnu-mathits/ABKT. is proposed, which introduces the ability factor into learning feedback attribution to enable the model to analyze the learning process from two perspectives, knowledge and ability, simultaneously. Based on constructive learning theory, continuous matrix factorization (CMF) model is proposed to simulate the knowledge internalization process, following the initiative growth and stationarity principles. In addition, the linear graph latent ability (LGLA) model is proposed to construct learner and item latent ability features, from graph-structured learner interaction data. Then, the knowledge and ability dual-tracing framework is constructed to integrate the knowledge and ability modules. Experimental results on four public databases indicate that the proposed methods perform better than state-of-the-art knowledge tracing algorithms in terms of prediction accuracy in quantitative assessments, displaying some advantages in model interpretability and intelligibility. [ABSTRACT FROM AUTHOR]

Feature importance and interaction are among the main issues in explainable artificial intelligence or interpretable machine learning. To measure feature importance and interaction, several methods, such as H-statistic and partial dependency, have been proposed. However, it is difficult to understand the practical implications of importance and interaction. In this paper, a new method for measuring feature importance and interaction is proposed. For the classification model, we observed correctly predicted cases in a predictive model and grouped them according to the characteristics of the cases. We derived a method for feature importance and interaction from group information. For the regression model, we grouped cases according to the change in the size of the prediction error. The proposed method supports the same rationale for feature importance and interaction. It also supports the decomposition of feature importance to feature power and feature interactions. To implement the proposed method, three visualization tools, including a feature interaction graph, are implemented. Through the proposed work, we can better understand the working mechanism of a predictive model. [ABSTRACT FROM AUTHOR]

The rapid integration of intelligent processes and methods into information systems in the Artificial Intelligence (AI) era has led to a substantial shift towards autonomous software decision-making. This evolution necessitates robust human oversight, especially in critical domains like Healthcare, Education, and Energy. Human trust in AI plays a vital role in influencing decision-making processes of users interacting with AI. This paper presents VIRTSI (V ariability and I mpact of R eciprocal T rust S tates towards I ntelligent systems), a novel rigorous computational model for human-AI Interaction. VIRTSI simulates human trust states, spanning from overtrust to distrust, through user modelling. It comprises: 1. A trust dynamics representational model based on Deterministic Finite State Automata (DFAs), illustrating transitions among cognitive trust states in response to AI-generated replies. 2. A trust evaluation model based on Confusion Matrices, originating from machine learning and Accuracy Metrics, providing a quantitative framework for analysing human trust dynamics. As a result, this is the first time that trust dynamics have been thoroughly traced in a representational model and a method has been developed to assess the impact of possibly harmful states like overtrust and distrust. An empirical study on the recently launched Large Language Model of generative AI, ChatGPT (version 3.5), provides a radical underexplored AI-generated platform for evaluating the human-AI interaction through VIRTSI. The study involved 1200 interactions of real users as well as AI experts together with experts in two very different domains of evaluation, namely software engineering and poetry. This study traces trust dynamics and the emerging human-AI interaction, in concrete examples of real user synergies with generative AI. The research reveals the vital role of maintaining normal trust states for optimal human-AI interaction and that both AI and human users need further steps towards this goal. The real-world implications of this research can guide the creation and evaluation of user interfaces with AI and the incorporation of functionalities in the development of generative AI chatbots in terms of trust by providing a new rigorous DFA representational method of trust dynamics and a corresponding new perspective of confusion matrix evaluation method of the dynamics' impact in the efficiency of human-AI dialogues. [ABSTRACT FROM AUTHOR]

The use of artificial intelligence for decision support and automation has shown tremendous potential in many areas. The ability to explain the decisions made by a machine learning algorithm is fundamental to facilitating the widespread use of this type of tool. There are many important real-world problems where the cost of the decisions depends on the characteristics of each example: these are called example-dependent cost (EDC) problems. For this type of classification problem, an appropriate formulation that takes into account the decision costs is fundamental both for the design of the classifier and for the explanation of its decisions. In this paper, we propose COCOA, an explanation method designed for EDC problems based on a Bayesian discriminant. The proposed method can provide counterfactual samples generated by considering decision costs. The COCOA method provides valid and plausible counterfactuals with a high success rate, which can be actionable, diverse, and sparse, achieving a remarkable improvement in terms of cost over five state-of-the-art methods on six real-world datasets. • COCOA provides actionable, plausible, sparse and diverse counterfactuals. • Decision costs are taken into account when generating counterfactuals. • The obtained counterfactuals improve cost savings. • The method is presented for MLPs, but other architectures can be used. [ABSTRACT FROM AUTHOR]

• Introduce a new algorithm for Gradient Boosting Decision Trees. • The method allows the exact sequence of decisions of the ensemble to be calculated. • This method allows the thresholds and features of each decision to be obtained. Gradient Boost Decision Trees (GBDT) is a powerful additive model based on tree ensembles. Its nature makes GBDT a black-box model even though there are multiple explainable artificial intelligence (XAI) models obtaining information by reinterpreting the model globally and locally. Each tree of the ensemble is a transparent model itself but the final outcome is the result of a sum of these trees and it is not easy to clarify. In this paper, a feature contribution method for GBDT is developed. The proposed method takes advantage of the GBDT architecture to calculate the contribution of each feature using the residue of each node. This algorithm allows to calculate the sequence of node decisions given a prediction. Theoretical proofs and multiple experiments have been carried out to demonstrate the performance of our method which is not only a local explicability model for the GBDT algorithm but also a unique option that reflects GBDTs internal behavior. The proposal is aligned to the contribution of characteristics having impact in some artificial intelligence problems such as ethical analysis of Artificial Intelligence (AI) and comply with the new European laws such as the General Data Protection Regulation (GDPR) about the right to explain and nondiscrimination. [ABSTRACT FROM AUTHOR]

In this work, with the help of a calculus system syllogistic logic with Carroll's diagrams (SLCD), we construct a useful algorithm for the possible deductions of polysyllogisms (soriteses). This algorithm makes a general deduction in categorical syllogisms with the help of diagrams to depict each proposition of polysyllogisms. The developed calculus system PolySLCD (PSLCD) is used to allow a formal deduction from premises set by comprising synchronically biliteral and triliteral diagrammatical appearance and simple algorithmic nature. This algorithm can be used to deduce new conclusions, step by step, through recursive conclusion sets that are obtained from premises of categorical polysyllogisms. The fundamental contributions of this paper are accurately deducing conclusions from sets corresponding to given premises as exact human reasoning using a single algorithm and designing this algorithm based on SLCD. Therefore, it is more suitable for computer-aided solution. Since the algorithm is set-based, it is a novel algorithm in the literature and it can easily contribute to the researchers using polysyllogisms in different scientific branches, such as computer science, decision-making systems and artificial intelligence. [ABSTRACT FROM AUTHOR]

• The stochastic separation theorems are applied to develop industrial computer vision systems. • New class of algorithms are presented and verified for removing AI errors. • This class of algorithms is based on stochastic separation theorems combined with clustering. • The algorithms are tested on a relevant real-life problem. In this paper we present theory and algorithms enabling classes of Artificial Intelligence (AI) systems to continuously and incrementally improve with a priori quantifiable guarantees – or more specifically remove classification errors – over time. This is distinct from state-of-the-art machine learning, AI, and software approaches. The theory enables building few-shot AI correction algorithms and provides conditions justifying their successful application. Another feature of this approach is that, in the supervised setting, the computational complexity of training is linear in the number of training samples. At the time of classification, the computational complexity is bounded by few inner product calculations. Moreover, the implementation is shown to be very scalable. This makes it viable for deployment in applications where computational power and memory are limited, such as embedded environments. It enables the possibility for fast on-line optimisation using improved training samples. The approach is based on the concentration of measure effects and stochastic separation theorems and is illustrated with an example on the identification faulty processes in Computer Numerical Control (CNC) milling and with a case study on adaptive removal of false positives in an industrial video surveillance and analytics system. [ABSTRACT FROM AUTHOR]

Progress in the field of Non-Intrusive Load Monitoring (NILM) has been attributed to the rise in the application of artificial intelligence. Nevertheless, the ability of energy disaggregation algorithms to disaggregate different appliance signatures from aggregated smart grid data poses some privacy issues. This paper introduces a new notion of disclosure risk termed energy disaggregation risk. The performance of Sequence-to-Sequence (Seq2Seq) NILM deep learning algorithm along with three activation extraction methods are studied using two publicly available datasets. To understand the extent of disclosure, we study three inference attacks on aggregated data. The results show that Variance Sensitive Thresholding (VST) event detection method outperformed the other two methods in revealing households' lifestyles based on the signature of the appliances. To reduce energy disaggregation risk, we investigate the performance of two privacy-preserving mechanisms based on microaggregation and Discrete Fourier Transform (DFT). Empirically, for the first scenario of inference attack on UK-DALE, VST produces disaggregation risks of 99%, 100%, 89% and 99% for fridge, dish washer, microwave, and kettle respectively. For washing machine, Activation Time Extraction (ATE) method produces a disaggregation risk of 87%. We obtain similar results for other inference attack scenarios and the risk reduces using the two privacy-protection mechanisms. [ABSTRACT FROM AUTHOR]

The shift to a new era of dealing with big data has driven continuous progress and development in computer science, artificial intelligence and machine learning. This change has led to the application of advanced techniques in the realm of decision science, particularly in the area of large-scale group decision-making (LSGDM). However, although these existing techniques have become the core of LSGDM methods, they are still limited in solving problems facing incomplete data. In addition, due to the rise of social media platforms such as Weibo, WeChat and Twitter, which build bridges for communication between decision makers (DMs), this brings new opportunities and challenges for consensus research. To address this set of issues, this study develops a consensus architecture that combines dynamic social network and opinion evolution in the context of an incomplete multi-attribute LSGDM. It is worth mentioning that the proposed consensus framework is a novel decision-making system that can be used to complete the estimation of the missing values and the consensus reaching process (CRP) by simulating the realistic decision-making scenarios. Firstly, considering the size of the trust value and the length of the path, a new trust propagation method is designed to achieve a more reliable estimation of the unknown trust value. Secondly, this paper establishes a missing value estimation method by virtue of the improved DeGroot model, which is able to obtain complete evaluation information by simulating the opinion formation process of DMs. Next, a hierarchical clustering algorithm with stronger robustness is constructed, which not only can adaptively complete the clustering process, but also integrally considers two attributes of trust and opinion similarity. In light of the above research, this study designs an opinion evolution and dynamic trust-driven consensus model, referred to as the DSN-DG-LSGDM model. Finally, the sensitivity analysis and experiments on a real dataset verify the significant superiority of the constructed DSN-DG-LSGDM model compared with the extant LSGDM consensus models. • A DeGroot model based on self-confidence level is designed. • An opinion evolution and dynamic trust-driven consensus framework is developed. • An adaptive clustering algorithm with dependence levels is constructed. • A more reliable trust propagation method is proposed. • A new dynamic subgroup weight calculation method is introduced. [ABSTRACT FROM AUTHOR]