Applied Intelligence

Zhang, Hong-Bo; Dong, Li-Jia; Lei, Qing; Yang, Li-Jie; Du, Ji-Xiang

doi: 10.1007/s10489-022-03984-5pmid: 35991679

Most existing action quality assessment (AQA) methods provide only an overall quality score for the input video and lack an evaluation of each substage of the movement process; thus, these methods cannot provide detailed feedback for users. Moreover, the existing datasets do not provide labels for substage quality assessment. To address these problems, in this work, a new label-reconstruction-based pseudo-subscore learning (PSL) method is proposed for AQA in sporting events. In the proposed method, the overall score of an action is not only regarded as a quality label but also used as a feature of the training set. A label-reconstruction-based learning algorithm is built to generate pseudo-subscore labels for the training set. Moreover, based on the pseudo-subscore labels and overall score labels, a multi-substage AQA model is fine-tuned from the PSL model to predict the action quality score of each substage and the overall score for an athlete. Several ablation experiments are performed to verify the effectiveness of each module. The experimental results show that our approach achieves state-of-the-art performance.

Liu, Peng; Qian, Wenhua; Cao, Jinde; Xu, Dan

doi: 10.1007/s10489-022-04012-2pmid: N/A

Deep learning models have achieved remarkable success in medical imaging analysis. However, existing methods are primarily focused on supervised learning, which requires a massive amount of training data. Recent studies have explored semi-supervised learning approaches to address this issue, where data augmentation was applied to unlabeled data. However, there are still two unsolved challenges in applying data augmentation to unlabeled medical images: it can i) result in the lesion features loss and ii) reduce the discriminability of prediction results. Thus, in this work, weak data augmentation is applied to unlabeled data to avoid losing lesions features. Also, we propose nuclear-norm maximization to achieve entropy minimization without losing prediction diversity. Experimental results on two public datasets show that the proposed method outperforms the compared models.

Shukla, Suyash; Kumar, Sandeep

doi: 10.1007/s10489-022-04002-4pmid: N/A

Software Effort Estimation (SEE) is a procedure to estimate the effort required to develop software. The researchers have been dealing with SEE issues for a long time. Several methods were developed until the formulation of Function Point (FP) and Constructive Cost Estimation (COCOMO) methods. However, these methods were useful only for procedurally developed software, not for modern object-oriented software. On the other hand, using the Use Case Point (UCP) metric acquired from the UML diagrams can be more suitable, as the use case is the fundamental unit of an object-oriented system. An ample amount of research has already been done for UCP prediction using linear regression-based models. However, various nonlinear regression models have not been explored for predicting UCP values from different UCP parameters. Although, some of the researchers have used nonlinear regression models for predicting effort, given the UCP value. Motivated by this, the current work investigates different nonlinear regression models such as a k-nearest neighbor, decision tree, random forest, support vector machine, and multilayer perceptron for UCP prediction. The experimental investigation has been conducted on two publicly available UCP estimation datasets. Further, we compared the performance of nonlinear regression models with the linear regression-based models using different performance measures. The results suggest that the nonlinear regression models perform better than the linear regression-based models.

Kashani, Elham S.; Bagheri Shouraki, Saeed; Norouzi, Yaser; De Baets, Bernard

doi: 10.1007/s10489-022-03711-0pmid: N/A

In this paper, we propose a novel density-grid-based method for clustering k-dimensional data. KIDS, an acronym for K-dimensional Ink Drop Spread, detects densely-connected pieces of data in k-dimensional grids. It enables one to simultaneously exploit the advantages of fuzzy logic, as well as both density-based and grid-based clustering. In the proposed method, the k-dimensional data space is divided into different cells. Input data records are mapped to the cells. The data points are then spread in the k-dimensional cells, just like what happens to ink drops in water. So the cells adjacent to the data cells also represent the data. Eventually, the impacts of all data grid cells are condensed and compared with the threshold to compute the final clusters. The experimental results show that the method has superior quality and efficiency in both low and high dimensions. In addition, the method is not only robust to noise but it is also capable of finding clusters of arbitrary shapes.

Ma, Rui; Du, Xian

doi: 10.1007/s10489-022-04068-0pmid: N/A

Advancement of imaging techniques enables consecutive image sequences to be acquired for quality monitoring of manufacturing production lines. Registration for these image sequences is essential for in-line pattern inspection and metrology, e.g., in the printing process of flexible electronics. However, conventional image registration algorithms cannot produce accurate results when the images contain duplicate and deformable patterns in the manufacturing process. Such a failure originates from the fact that the conventional algorithms only use spatial and pixel intensity information for registration. Considering the nature of temporal continuity of the product images, in this paper, we propose a closed-loop feedback registration algorithm. The algorithm leverages the temporal and spatial relationships of the consecutive images for fast, accurate, and robust point matching. The experimental results show that our algorithm finds about 100% more matching point pairs with a lower root mean squared error and reduces up to 86.5% of the running time compared to other state-of-the-art outlier removal algorithms.

Xu, Dong-qin; Li, Ming-ai

doi: 10.1007/s10489-022-04077-zpmid: 36039116

Domain adaptation, as an important branch of transfer learning, can be applied to cope with data insufficiency and high subject variabilities in motor imagery electroencephalogram (MI-EEG) based brain-computer interfaces. The existing methods generally focus on aligning data and feature distribution; however, aligning each source domain with the informative samples of the target domain and seeking the most appropriate source domains to enhance the classification effect has not been considered. In this paper, we propose a dual alignment-based multi-source domain adaptation framework, denoted DAMSDAF. Based on continuous wavelet transform, all channels of MI-EEG signals are converted respectively and the generated time-frequency spectrum images are stitched to construct multi-source domains and target domain. Then, the informative samples close to the decision boundary are found in the target domain by using entropy, and they are employed to align and reassign each source domain with normalized mutual information. Furthermore, a multi-branch deep network (MBDN) is designed, and the maximum mean discrepancy is embedded in each branch to realign the specific feature distribution. Each branch is separately trained by an aligned source domain, and all the single branch transfer accuracies are arranged in descending order and utilized for weighted prediction of MBDN. Therefore, the most suitable number of source domains with top weights can be automatically determined. Extensive experiments are conducted based on 3 public MI-EEG datasets. DAMSDAF achieves the classification accuracies of 92.56%, 69.45% and 89.57%, and the statistical analysis is performed by the kappa value and t-test. Experimental results show that DAMSDAF significantly improves the transfer effects compared to the present methods, indicating that dual alignment can sufficiently use the different weighted samples and even source domains at different levels as well as realizing optimal selection of multi-source domains.

Ding, Xinyi; Han, Tao; Fang, Yili; Larson, Eric

doi: 10.1007/s10489-022-04095-xpmid: N/A

Knowledge Tracing is the process of tracking mastery level of different skills of students for a given learning domain. It is one of the key components for building adaptive learning systems and has been investigated for decades. The empirical success of deep neural networks in the past few years has encouraged researchers in the learning science community to take similar approaches. However, most existing deep learning based knowledge tracing models have the following limitations: (1) Only use the correct/incorrect response, ignoring useful information from other modalities. (2) For works do consider multimodality use simple concatenation, which might not be the best way for modality fusion. (3) Design their network architectures manually via trial and error. To solve these problems, we propose Multimodal Fusion and Neural Architecture Search (MFNAS) in this paper. The commonly used neural architecture search technique could be considered as a special case of our proposed approach when there is only one modality involved. We further propose to use a new metric called time-weighted Area Under the Curve (weighted AUC) to measure how a sequence model performs with time. Our proposed approach MFNAS allows more efficient design of Knowledge Tracing models. Besides, evaluated on two public real datasets, the discovered model is able to achieve around 12% improvement in coefficient of determination, 2% improvement in AUC and weighted AUC compared with state of the art models. What’s more, unlike most existing works, we conduct McNemar’s test on the model predictions and the results are statistically significant.

Yildirim, Gungor

doi: 10.1007/s10489-022-03972-9pmid: 36068811

In fake news detection, intelligent optimization seems to be a more effective and explainable solution methodology than the black-box methods that have been extensively used in the literature. This study takes the optimization-based method one step further and proposes a novel, multi-thread hybrid metaheuristic approach for fake news detection in social media. The most innovative feature of the proposed method is that it uses a supervisor thread mechanism, which simultaneously monitors and improves the performance and search patterns of metaheuristic algorithms running parallel. With the supervisor thread mechanism, it is possible to analyse different key attribute combinations in the search space. In addition, this study develops a software framework that allows this model to be implemented easily. It tests the performance of the proposed model on three different data sets, respectively containing news about Covid-19, the Syrian War, and daily politics. The proposed method is evaluated in comparison to the results of fifteen different well-known deep models and classification algorithms. Experimental results prove the success of the proposed model and that it can produce competitive results.