Abstract: Random cut trees are generated with respective to respective samples of a baseline set of data records of a data set for which outlier detection is to be performed. To construct a particular random cut tree, an iterative splitting technique is used, in which the attribute along which a given set of data records is split is selected based on its value range. With respect to a newly-received data record of the stream, an outlier score is determined based at least partly on a potential insertion location of a node representing the data record in a particular random cut tree, without necessarily modifying the random cut tree.

Abstract: Examples of techniques for anomaly detection with feedback are described. An instance includes a technique is receiving a plurality of unlabeled data points from an input stream; performing anomaly detection on a point of the unlabeled data points using an anomaly detection engine; pre-processing the unlabeled data point that was subjected to anomaly detection; classifying the pre-processed unlabeled data point; determining the anomaly detection was not proper based on a comparison of a result of the anomaly detection and a result of the classifying of the pre-processed unlabeled data point; and in response to determining the anomaly detection was not proper, providing feedback to the anomaly detection engine to change at least one emphasis used in anomaly detection.

Abstract: Random cut trees are generated with respective to respective samples of a baseline set of data records of a data set for which outlier detection is to be performed. To construct a particular random cut tree, an iterative splitting technique is used, in which the attribute along which a given set of data records is split is selected based on its value range. With respect to a newly-received data record of the stream, an outlier score is determined based at least partly on a potential insertion location of a node representing the data record in a particular random cut tree, without necessarily modifying the random cut tree.

Abstract: Random cut trees are generated with respective to respective samples of a baseline set of data records of a data set for which outlier detection is to be performed. To construct a particular random cut tree, an iterative splitting technique is used, in which the attribute along which a given set of data records is split is selected based on its value range. With respect to a newly-received data record of the stream, an outlier score is determined based at least partly on a potential insertion location of a node representing the data record in a particular random cut tree, without necessarily modifying the random cut tree.

Abstract: Techniques for detecting anomalies in streaming graph data are described. For example, an embedding technique of generating a multi-dimensional vector of summations of each weighted edge found in both a random source bounding proper subset and a random destination bounding proper subset associated with a dimension of the epoch graph is detailed. Anomaly detection is performed on the generated multi-dimensional vectors.

Abstract: Techniques for seasonality-based anomaly detection and forecast are described. For example, a method of receiving a request to generate forecast for received time series data; performing a seasonality-based anomaly detection and forecast for the received time series data based upon the received request, the seasonality-based anomaly detection and forecasting to utilize a second data structure that reflect anomalies found in a first data structure on the input from the received time series data; and providing a result of the performed seasonality-based anomaly detection and forecast is described.

Abstract: At an artificial intelligence system, a random cut tree corresponding to a sample of a multi-dimensional data set is traversed to determine a tree-specific vector indicating respective contributions of individual dimensions to an anomaly score of a particular data point. Level-specific vectors of per-dimension contributions obtained using bounding-box analyses at each level during the traversal are aggregated to obtain the tree-specific vector. An overall anomaly score contribution for at least one dimension is obtained using respective tree-specific vectors generated from one or more random cut trees, and an indication of the overall anomaly score contribution is provided.

Abstract: A stacked-chip assembly including a plurality of IC chips or die that are stacked, and electrically coupled by solder bonds. In accordance with some embodiments described further below, the solder bonds are to contact a back-side land that includes a diffusion barrier to reduce intermetallic formation and/or other solder-induced reliability issues. The back-side land may include an electrolytic nickel (Ni) barrier layer separating solder from a back-side redistribution layer trace. This electrolytic Ni may be of high purity, which at least in part, may enable the backside metallization stack to be of minimal thickness while still functioning as a diffusion barrier. In some embodiments, the back-side land composition and architecture is distinct from a front-side land composition and/or architecture.

Abstract: Random cut trees are generated with respective to respective samples of a baseline set of data records of a data set for which outlier detection is to be performed. To construct a particular random cut tree, an iterative splitting technique is used, in which the attribute along which a given set of data records is split is selected based on its value range. With respect to a newly-received data record of the stream, an outlier score is determined based at least partly on a potential insertion location of a node representing the data record in a particular random cut tree, without necessarily modifying the random cut tree.

Abstract: Certain exemplary embodiments provide a method comprising: automatically: receiving a plurality of elements for each of a plurality of continuous data streams; treating the plurality of elements as a first data stream matrix that defines a first dimensionality; reducing the first dimensionality of the first data stream matrix to obtain a second data stream matrix; computing a singular value decomposition of the second data stream matrix; and based on the singular value decomposition of the second data stream matrix, quantifying approximate linear correlations between the plurality of elements.

Abstract: A system and method are provided for summarizing dynamic data from distributed sources through the use of histograms. In particular, the method comprises receiving a first data signal at a first location, determining a first array sketch of the first data signal, and constructing a first output histogram from the first array sketch and a first robust histogram via a first hybrid histogram. Array sketches of a number of data signals may be calculated, and added to yield a single vector sum. The histogram is constructed from the vector sum. In that way, the vector sum may be analyzed without revealing the individual data signals that form the basis of the sum.

Abstract: The present invention relates to a method and apparatus for optimizing queries. The present invention discloses an efficient method for providing answers to queries under parametric aggregation constraints.

Abstract: The present invention relates to a method and apparatus for optimizing queries. The present invention discloses an efficient method for providing answers to queries under parametric aggregation constraints.

Abstract: The present invention relates to a method and apparatus for optimizing queries. The present invention discloses an efficient method for providing answers to queries under parametric aggregation constraints.

Abstract: Certain exemplary embodiments provide a method comprising: automatically: receiving a plurality of elements for each of a plurality of continuous data streams; treating the plurality of elements as a first data stream matrix that defines a first dimensionality; reducing the first dimensionality of the first data stream matrix to obtain a second data stream matrix; computing a singular value decomposition of the second data stream matrix; and based on the singular value decomposition of the second data stream matrix, quantifying approximate linear correlations between the plurality of elements.

Abstract: A device and a method are provided. Approximate match operations are performed for each of a group of attributes for each of a group of tuples with respect to a query to create a respective ranking for each of the group of attributes. The rankings of the group of attributes are combined to provide a ranking score for each of the group of tuples. Data representing a ranking score of each of the group of tuples is generated according to a position of a respective ranking of each one of the group of tuples for a first k positions of the ranking. K of top ranked ones of the group of tuples are identified based at least in part on the generated data, wherein a number of the group of tuples is n and k<n.

Abstract: The present invention relates to a method and apparatus for optimizing queries. The present invention discloses an efficient method for providing answers to queries under parametric aggregation constraints.

Abstract: A system and method are provided for monitoring dynamic data from distributed sources through the use of histograms. In the method, an array sketch of the digital signal is determined, a robust histogram is constructed from the array sketch, and an output histogram is constructed from the array sketch and the robust histogram via a hybrid histogram. Dyadic intervals of a representation of the array sketch are used in constructing the robust histogram.

Abstract: A technique that uses a weighted divide and conquer approach for clustering a set S of n data points to find k final centers. The technique comprises 1) partitioning the set S into P disjoint pieces S1, . . . , Sp; 2) for each piece Si, determining a set Di of k intermediate centers; 3) assigning each data point in each piece Si to the nearest one of the k intermediate centers; 4) weighting each of the k intermediate centers in each set Di by the number of points in the corresponding piece Si assigned to that center; and 5) clustering the weighted intermediate centers together to find said k final centers, the clustering performed using a specific error metric and a clustering method A.

Abstract: A technique that uses a weighted divide and conquer approach for clustering a set S of n data points to find k final centers. The technique comprises 1) partitioning the set S into P disjoint pieces S1, . . . , Sp; 2) for each piece Si, determining a set Di of k intermediate centers; 3) assigning each data point in each piece Si to the nearest one of the k intermediate centers; 4) weighting each of the k intermediate centers in each set Di by the number of points in the corresponding piece Si assigned to that center; and 5) clustering the weighted intermediate centers together to find said k final centers, the clustering performed using a specific error metric and a clustering method A.