44 results about "Biclustering" patented technology

The invention discloses an order-preserving submatrix (OPSM) and frequent sequence mining based emotion classification method for e-commerce comments. The method comprises the following steps: (1) performing pretreatment and Chinese word segmentation on the e-commerce comments; calculating to obtain a TF-IDF weight vector of synonyms; and then mining a local mode in the weight vector through a biclustering algorithm based on OPSM; (3) mining classification frequent phrase characteristics through an improved PrefixSpan algorithm, and meanwhile, improving the capacity for distinguishing emotion tendency by the frequent phrases through limitation such as word intervals; and (4) converting the characteristics mined in steps (2) and (3) into a 0 / 1 vector to be used as an input of a classifier, and thus obtaining the emotion classification result of the e-commerce comments. With the adoption of the method, the emotion classification characteristics of the e-commerce comments can be accurately mined, so that potential customers can know the goods evaluation information before buying, and meanwhile, the businessman can fully know the suggestions of the customers and accordingly improve the service quality.

Methods and systems for improved unsupervised learning are described. The unsupervised learning can consist of biclustering a data set, e.g., by biclustering subsets of the entire data set. In an example, the biclustering does not include feeding know and proven results into the biclustering methodology or system. A hierarchical approach can be used that feeds proven clusters back into the biclustering methodology or system as the input. Data that does not cluster may be discarded. Thus, a very large unknown data set can be acted on to learn about the data. The system is also amenable to parallelization.

The invention belongs to the field of indoor positioning technology, and discloses an indoor positioning method based on CSI space-frequency characteristics and a reference point location clustering algorithm. According to the method, CSI data of which a physical layer is finer-grained and more robust is taken as physical information to ensure that the perceptual ability of the system is much higher than that of RSS, and the influence of multi-path effects can be effectively avoided; CSI space-frequency characteristic vector clustering and reference point location multi-index clustering technologies are adopted, members of which the geographical locations are relatively far away from other cluster members in an original cluster are separately divided into a cluster, and thus the phenomenon that reference points which have relatively far geographical locations and approximate CSI space-frequency characteristic vectors are merged into the same cluster can be avoided; and a biclustering technology including CSI space-frequency characteristic clustering and reference point location clustering is adopted, and the clusters that should not be split are merged into a cluster again, the integrity of the original clusters can be ensured while the processing of singular points is completed, and thus a better clustering effect can be achieved.

The invention discloses a double-clustering method for tumor gene expression profile data based on double hypergraph regularization, by clustering the samples and genes of the tumor gene expression profile data respectively; then, the samples and genes of the tumor gene expression profile data Genes respectively establish sample hypergraph and gene hypergraph to obtain the inherent geometric structure of samples and genes; finally, the sample hypergraph and gene hypergraph are respectively used as the sample hypergraph regularization item and gene hypergraph regularization item of principal component analysis, and determine The objective function is optimized, and finally the sample clustering matrix and gene clustering matrix are respectively optimized by optimizing the objective function to obtain the final sample clustering and gene clustering. Based on the principle component analysis method, the present invention optimizes the double clustering through double hypergraph regularization, so as to better obtain the complex information in the tumor gene expression profile data on the basis of retaining the advantages of the principal component analysis method, Finally, the accuracy of clustering is improved.

The invention discloses a double-clustering method of gene expression data based on double-stage genetic calculation, and subtracts the k-th column from each column in the matrix M to obtain the matrix M(k), k=1, 2,...,n; Perform hierarchical clustering on each column of M(k) to obtain a set of bi-clustering seeds; obtain the corresponding bi-clustering through genetic calculation. The algorithm described in the present invention can solve the shortcoming that the traditional biclustering algorithm based on genetic calculation can only select for biclustering, by optimizing the ranks and columns at the same time, the search efficiency can be improved, and a better biclustering analysis can be obtained Effect.

Systems, methods, and non-transitory computer-readable media (systems) are disclosed for generating meaningful and insightful user segment reports based on a high dimensional data space. In particular, in one or more embodiments, the disclosed systems utilize a relaxed bi-clustering model to automatically identify user segments in a data space including datasets of features specific to individual users. In at least one embodiment, the disclosed systems identify and include users in automatically generated user segments even though those users are associated with some, but perhaps not all, of the features as other members in the automatically generated user segments.