Abstract: Methods, systems, and computer-readable and executable instructions are provided for classifying an electrocardiogram (ECG) signal. Classifying an ECG signal can include analyzing the ECG signal using a stream of pulses generated by a sampler, extracting cardiac pulse features from a timing of the stream of pulses, and classifying the ECG signal based on the extracted cardiac pulse feature.

Abstract: Systems and methods of anomaly detection in data centers. An example method may include analyzing time series data for the data center by testing statistical hypotheses. The method may also include constructing upper and lower bounds based on the statistical hypotheses. The method may also include flagging anomalies in the time series data falling outside of the upper and lower bounds.

Abstract: A system and method for classifying unclassified samples. The method includes detecting a number of classes including training samples in training data sets. The method includes, for each class, determining a vector for each training sample based on a specified number of nearest neighbor distances between the training sample and neighbor training samples, and determining a class distribution based on the vectors. The method also includes detecting an unclassified sample in a data set and, for each class, determining a vector for the unclassified sample based on the specified number of nearest neighbor distances between the unclassified sample and nearest neighbor training samples within the class, and determining a probability that the unclassified sample is a member of the class based on the vector and the class distribution. The method further includes classifying the unclassified sample based on the probabilities.

Abstract: Systems and methods for multi-regime detection in streaming data are disclosed. An example method includes generating vectors for each sample of the streaming data. The method also includes inducing mean independence of the vectors to find an embedded data trajectory. The method also includes comparing the embedded data trajectory with known data trajectories. The method also includes issuing an alert if the embedded data trajectory corresponds to a known data trajectory indicating an anomaly in the streaming data.

Abstract: A method and apparatus are disclosed for identifying anomalies of a signal, by analyzing a signal using a frequency-based technique, analyzing results of the frequency-based analysis using a statistical analysis technique, determining one or more limits based on the statistical analysis, and comparing a frequency domain representation of the signal to the limits to identify anomalies of the signal.

Abstract: Estimation of unique values in a database can be performed where a data field having multiple information values is provided in the database. The data field can be partitioned into multiple intervals such that each interval includes a range of information values. An interval specific Bloom filter can be calculated for each of the multiple intervals. A binary Bloom filter value can be calculated for an information value within an interval specific Bloom filter. The binary Bloom filter value can represent whether the information value is unique. A number of unique values in the database can be determined based on calculated binary Bloom filter values.

Abstract: The present disclosure relates to computing techniques. Data arrays are processed using Bloom filters to determine aggregate count, maximum, and minimum. These methods can be used on different types of data, including data groups, partial groups, data cubes, hypercubes, and others.

Abstract: The present disclosure generally relates to data processing. Bloom filters are used to process data at high speed. A Bloom filter that is initialized based on a source string can be used to quickly determine the similarity between the source string and a query string.

Abstract: Compression of an initial dataset is implemented on a data processing system. The initial dataset can be transformed (210) into a group of initial wavelet coefficients using a wavelet basis function. Magnitudes of initial wavelet coefficients in the group of initial wavelet coefficients can be calculated (220). Initial wavelet coefficients having magnitudes beyond a cutoff value can be deleted (230). A compressed group of wavelet coefficients can be identified (240) from the wavelet coefficients remaining within the cutoff value. The initial dataset can be approximated (250) using the compressed group of wavelet coefficients and the wavelet basis function.

Abstract: Probable anomalies associated with at least one data metric may be detected across a series of windows of time series data by comparison of data to a threshold. An estimated probability of anomalies for each of the windows of time series data may be determined based on the detected probable anomalies and the threshold. The windows of time series data may be ranked based on the estimated probabilities. Probable anomalies associated with highest ranked windows of time series data may be output to a user.

Abstract: Systems and methods for anomaly detection in streaming data are disclosed. An example method includes applying statistical analysis to streaming data in a sliding window. The method also includes extracting a feature. The method also includes determining class assignment for the feature using class conditional probability densities and a threshold.

Abstract: Systems and methods of anomaly detection in data centers. An example method may include analyzing time series data for the data center by testing statistical hypotheses. The method may also include constructing upper and lower bounds based on the statistical hypotheses. The method may also include flagging anomalies in the time series data falling outside of the upper and lower bounds.

Abstract: Systems and methods for multi-regime detection in streaming data are disclosed. An example method includes generating vectors for each sample of the streaming data. The method also includes inducing mean independence of the vectors to find an embedded data trajectory. The method also includes comparing the embedded data trajectory with known data trajectories. The method also includes issuing an alert if the embedded data trajectory corresponds to a known data trajectory indicating an anomaly in the streaming data.

Abstract: Systems and methods for detecting anomalies in a large scale and cloud datacenter are disclosed. Anomaly detection is performed in an automated, statistical-based manner by using a parametric Gini coefficient technique or a non-parametric Tukey technique. In the parametric Gini coefficient technique, sample data is collected within a look-back window. The sample data is normalized to generate normalized data, which is binned into a plurality of bins defined by bin indices. A Gini coefficient and a threshold are calculated for the look-back window and the Gini coefficient is compared to the threshold to detect an anomaly in the sample data. In the non-parametric Tukey technique, collected sample data is divided into quartiles and compared to adjustable Tukey thresholds to detect anomalies in the sample data.

Abstract: Systems and methods of time encoding using an integrate and fire (IF) sampler are disclosed. In an example, a method includes receiving input signals for separate classes. The method also includes generating a pulse train based on the input signals. The method also includes binning the pulse train to generate a feature vector.

Abstract: Methods, systems, and computer program products are provided for quantifying financial impact of marketing investments. Time series data describing the financial performance generated by corresponding marketing investments that are made as a function of time is provided to configure an econometric model. The econometric model, which describes a linear relationship between the financial performance and the corresponding marketing investments, is transformed into an aggregated non-linear econometric model that includes non-linear factors causing the financial performance to change at a varying rate as a function of the marketing investments.

Abstract: Methods, systems, and computer program products are provided for incorporating qualitative factors into an econometric model. Time series data describing the financial performance generated by corresponding marketing investments that are made as a function of time is provided to configure an econometric model. The econometric model includes linear coefficients that define a linear relationship between the financial performance and the corresponding marketing investments. The linear coefficients are adjusted in accordance with the qualitative factors received as inputs from experts, thereby enabling the qualitative factors to be quantified into the econometric model.

Abstract: A method determines a number of unique values in a sample of a list of values and estimates a number of the unique values for an unsampled portion of the list of values. The method estimates a number of the unique values in the list by adding the number of unique values in the sample to the number of the unique values in the unsampled portion.

Abstract: A method and apparatus are disclosed for identifying anomalies of a signal, by analyzing a signal using a frequency-based technique, analyzing results of the frequency-based analysis using a statistical analysis technique, determining one or more limits based on the statistical analysis, and comparing a frequency domain representation of the signal to the limits to identify anomalies of the signal.

Abstract: Estimation of unique values in a database can be performed where a data field having multiple information values is provided in the database. The data field can be partitioned into multiple intervals such that each interval includes a range of information values. An interval specific Bloom filter can be calculated for each of the multiple intervals. A binary Bloom filter value can be calculated for an information value within an interval specific Bloom filter. The binary Bloom filter value can represent whether the information value is unique. A number of unique values in the database can be determined based on calculated binary Bloom filter values.