Knowledge base‐based intelligent question‐answering systems have insufficient understanding of the questions. In the early stages of research, it is effective in most cases that the existing natural language question‐understanding methods can answer questions by connecting entities and relationships when ignoring the identification of focus words. However, as research deepens, ignoring focus words has become a shortcoming. To address this, we propose identifying focus words, enabling more precise understanding of user focus. We define focus itemset, frequent focus itemset, focus association rule, and strong focus association rule to express focus‐related information better. Given the unique nature of focus association rules, we propose a prefix tree structure and an algorithm for mining association rules aimed at identifying focus words. We also introduce an inverted index specifically designed for focus association rules and propose an efficient algorithm for identifying focus words based on this index. Experiments verify the effectiveness of our algorithm and the efficiency of the inverted index, with a focus word identification rate exceeding 90%.

Although spectral clustering is capable of identifying clusters of arbitrary shapes, its high time and space complexity poses limitations in large-scale data clustering applications. To tackle this problem, researchers have proposed using anchor points to construct the similarity matrix, thereby reducing time and space complexity. However, current methods for generating anchor points do not fit the data well and are limited in approach. To improve upon existing anchor points generation methods, we proposes a pseudo-label-based anchor points generation approach and develops a fast spectral clustering algorithm for large-scale data, named FSC-PLGB. The algorithm first randomly selects r points as an initial granular-ball, applies K-Means on these points to obtain pseudo-labels, calculates the pseudo-purity of the granular-ball based on these pseudo labels, and then performs granular-ball division based on these pseudo-purity to generate anchor points. A similarity matrix is constructed between all sample points and anchor points, and finally, spectral clustering is applied to obtain the clustering results. The experimental results demonstrate that our proposed algorithm exhibits exceptional efficiency and significant superiority on large-scale datasets. The source code is available at https://github.com/DongdongCheng/FSC-PLGB.

ABSTRACT The core challenge of fine‐grained image recognition (FGIR) tasks is distinguishing highly similar subclasses within the same base category. Most CNN‐based deep learning methods typically focus on extracting information from local regions while overlook the inherent structure between subclasses and the complex relationships between features. This paper presents a two‐stage feature selection method based on partial order analysis (POA) and heterogeneity evaluation (HE) for FGIR tasks, guiding the model to focus on distinctive features while reducing uncertainty caused by interfering information. Specifically, in the POA stage, clustering first groups similar subcategories into a medium‐granularity category. Formal concept analysis then models their hierarchical partial order, identifying “shared features” among subcategories and “exclusive features” unique to each. This structured representation highlights key contrastive cues. In the HE stage, a novel heterogeneity index is introduced to measure the fluctuation of low‐level features within each fine‐grained category. This index guides the model to suppress pseudo‐discriminative features with high heterogeneity, mitigating the impact of noisy and unstable information on decision‐making. We perform comprehensive experiments on three commonly used benchmark datasets (CUB‐200‐2011, Stanford Cars, and FGVC‐Aircraft). Experimental results show that the proposed method outperforms classic FGIC methods, validating the effectiveness of our approach.

In order to enhance the efficiency of regression testing, test case prioritization (TCP) has been widely implemented, wherein a higher priority test case is executed earlier. Traditional TCP methods focus on improving the prioritization algorithm's efficacy. However, the majority of TCP approaches are characterized by a predetermined sequence of test cases prior to execution. Once established, this sequence remains consistent throughout the entire test execution process. As a result, any execution information generated during current test execution (such as fault-detected information) is unavailable for use in current round of test case prioritization and can only be utilized in subsequent regression testing. To address the issue of lagging utilization of fault-detected information, a cluster-based adaptive test case prioritization approach is proposed, which adds the new adaptive adjustment content in pre-prioritization. First, a new clustering criterion is defined and designed, by which produces test-case clusters in advance. Second, an adaptive TCP algorithm is proposed, which utilizes fault-detected information to adaptively adjust the order of test cases during the execution process based on the test-case clusters. Finally, one open-source Java program and three industrial-grade Java programs were selected for empirical evaluation. The experimental results demonstrate that the proposed technique not only serves as an enhanced version of pre-prioritization to improve the performance of the corresponding pre-prioritization technique, but also functions as an independent approach that outperforms other TCP techniques, including cluster-based TCPs, and another adaptive TCP. Specifically, when step=2 is applied using our cluster-based adaptive TCP approach, the results are significantly better than those obtained with step=1. For instance, in CT-14 , the median APFD improvement rate for step=2 reaches 17.08 %, which is substantially higher than that achieved with step=1 (5.48 %).

Due to simplicity, K-means has become a widely used clustering method. However, its clustering result is seriously affected by the initial centers and the allocation strategy makes it hard to identify manifold clusters. Many improved K-means are proposed to accelerate it and improve the quality of initialize cluster centers, but few researchers pay attention to the shortcoming of K-means in discovering arbitrary-shaped clusters. Using graph distance (GD) to measure the dissimilarity between objects is a good way to solve this problem, but computing the GD is time-consuming. Inspired by the idea that granular ball uses a ball to represent the local data, we select representatives from a local neighborhood, called natural density peaks (NDPs). On the basis of NDPs, we propose a novel K-means algorithm for identifying arbitrary-shaped clusters, called NDP-Kmeans. It defines neighbor-based distance between NDPs and takes advantage of the neighbor-based distance to compute the GD between NDPs. Afterward, an improved K-means with high-quality initial centers and GD is used to cluster NDPs. Finally, each remaining object is assigned according to its representative. The experimental results show that our algorithms can not only recognize spherical clusters but also manifold clusters. Therefore, NDP-Kmeans has more advantages in detecting arbitrary-shaped clusters than other excellent algorithms.

Density peaks clustering algorithm (DP) has difficulty in clustering large-scale data, because it requires the distance matrix to compute the density and -distance for each object, which has time complexity. Granular ball (GB) is a coarse-grained representation of data. It is based on the fact that an object and its local neighbors have similar distribution and they have high possibility of belonging to the same class. It has been introduced into supervised learning by Xia et al. to improve the efficiency of supervised learning, such as support vector machine, -nearest neighbor classification, rough set, etc. Inspired by the idea of GB, we introduce it into unsupervised learning for the first time and propose a GB-based DP algorithm, called GB-DP. First, it generates GBs from the original data with an unsupervised partitioning method. Then, it defines the density of GBs, instead of the density of objects, according to the centers, radius, and distances between its members and centers, without setting any parameters. After that, it computes the distance between the centers of GBs as the distance between GBs and defines the -distance of GBs. Finally, it uses GBs' density and -distance to plot the decision graph, employs DP algorithm to cluster them, and expands the clustering result to the original data. Since there is no need to calculate the distance between any two objects and the number of GBs is far less than the scale of a data, it greatly reduces the running time of DP algorithm. By comparing with -means, ball -means, DP, DPC-KNN-PCA, FastDPeak, and DLORE-DP, GB-DP can get similar or even better clustering results in much less running time without setting any parameters. The source code is available at https://github.com/DongdongCheng/GB-DP.

Stacking pseudoinverse learner (SP) is an ensemble learning technology, and its generalization performance greatly affects the effect of image classification. Currently, most SPs randomly initialize the input weight matrix in a random subspace without limiting the random initial values, resulting in unstable training results and a decrease in generalization performance; in addition, training all samples at once may cause the classifier redundant and also affect the generalization performance of the model. To efficiently address the above issues, we propose a new framework called progressive stacking pseudoinverse learner (PSP), which aims to enhance the generalization performance of SP via active learning (AL) in random subspaces. Specifically, on the one hand, a random feature SP (RFSP) model is proposed, which constrains the random subspace by initializing the input weight matrix into different random specific distributions to improve the generalization performance of SP. On the other hand, an AL progressive (ALP) model based on RFSP is proposed. By iteratively selecting useful samples to optimize the classification results, the training sample information is effectively used to progressively enhance the generalization performance of the model. Experimental results on three public datasets show that our proposed PSP algorithm achieves better performance in accuracy, precision, recall, and$F1$score, and the results are competitive with state-of-the-art methods.

Outlier detection is one of the hot research in data mining , and has been applied to various fields such as network anomaly detection , image abnormal analysis, etc. In recent years, many outlier detecting algorithms have been proposed. However, these outlier detecting algorithms are hard to effectively detect global outliers, local outliers and outlier clusters at the same time. In this paper, we propose a novel outlier detecting algorithm based on the following ideas: (1) the density distribution should not be changed dramatically on local area; (2) the ratio of the number of k nearest neighbors and the number of reverse k nearest neighbors should not be very big. Based on above ideas, the proposed algorithm aims to find outlier turning points, then regards all outlier turning points and its sparse neighbors as outliers. Furthermore, the proposed algorithm use natural neighbors to obtain the neighborhood parameter k adaptively. The formal analysis and extensive experiments demonstrate that this technique can detect global outliers, local outliers and outlier clusters without neighborhood parameter k .

Spectral clustering is becoming more and more popular because it has good performance in discovering clusters with varying characteristics. However, it suffers from high computational cost, unstable clustering results and noises. This work presents a novel approximate spectral clustering based on dense cores and density peaks, called DCDP-ASC. It first finds a reduced data set by introducing the concept of dense cores; then defines a new distance based on the common neighborhood of dense cores and calculates geodesic distances between dense cores according to the new defined distance; after that constructs a decision graph with a parameter-free local density and geodesic distance for obtaining initial centers; finally calculates the similarity between dense cores with their new defined geodesic distance, employs normalized spectral clustering method to divide dense cores, and expands the result on dense cores to the whole data set by assigning each point to its representative. The results on some challenging data sets and the comparison of our algorithm with some other excellent methods demonstrate that the proposed method DCDP-ASC is more advantageous in identifying complex structured clusters containing a lot of noises.

In this paper, we propose a scalable mining algorithm to discover contextual outliers using relevant subspaces. We develop the mining algorithm using the MapReduce programming model running on a Hadoop cluster. Relevant subspaces, which effectively capture the local distribution of various datasets, are quantified using local sparseness of attribute dimensions. We design a novel way of calculating local outlier factors in a relevant subspace with the probability density of local datasets; this new approach can effectively reflect the outlier degree of a data object that does not satisfy the distribution of the local dataset in the relevant subspace. Attribute dimensions of a relevant subspace, and local outlier factors are expressed as vital contextual information, which improves the interpretability of outliers. Importantly, the selection of N data objects with the largest local outlier factor value is categorized as contextual outliers in our solution. To this end, our scalable mining algorithm, which incorporates the locality sensitive hashing distributed strategy, is implemented on a Hadoop cluster. The experimental results validate the effectiveness, interpretability, scalability, and extensibility of the algorithm using both synthetic data and stellar spectral data as experimental datasets.

Existing distance-based outlier mining methods do not consider the impact of each attribute's importance degree, thereby resulting in poor mining accuracies. To address this problem, we propose a new outlier mining algorithm – Miner* – that makes use of information entropy and Weighted Distance Sum to substantially improve mining accuracies. Miner* employs information entropy to determine weight values indicating the importance degrees of data attributes. An input dataset is reduced by Miner* through the neighbour-radius-based pruning technologies. Thus, Miner* obtains a candidate outlier set by removing any data objects that are unlikely to be outliers. Miner* calculates the weighted distance sum value Wkof each object in the candidate outlier set; Wkvalue ranks the top n to be regarded as outliers. Due to the sum of distance, which takes full advantage of the clustering characteristics of the dataset, edge distribution data objects and local outliers can be effectively mined out. To demonstrate the effectiveness of the Miner* algorithm, we implement Miner* in a prototype system to detect star spectrum data objects with abnormal characteristic lines. Our experimental results show that the algorithm in Miner* achieves high accuracy, high scalability, and low man-made influence by utilizing UCI and star spectrum dataset. Our results also confirm that Miner* is feasible and effective in mining spectrum data with abnormal characteristic lines from massive star spectrum dataset.