Abstract: A method includes obtaining, based on a sequential graph-based model, gaming exposure information over time, where the gaming exposure information includes device-level preferences and household-level preferences. The method also includes combining one or more raw user behavior sessions into a gameplay session based on the obtained gaming exposure information. The method further includes providing a scoring metric to (i) check an extent of multi-matching in the obtained gaming exposure information and (ii) remove untrustworthy gaming exposures from the obtained gaming exposure information. In addition, the method includes generating, based on a feature engineering pipeline, one or more game segments running in a production environment, where the one or more game segments are identified for ancillary content based on inferences by a machine learning model trained using the gaming exposure information.

Abstract: In one embodiment, a method includes receiving user behavior data and contextual information associated with the user behavior data, the contextual information including a first data portion associated with a first context type. The method includes generating, from the user behavior data and the contextual information using a hashing algorithm, a first heterogeneous hash code including a first portion representing the user behavior data and a second hash code portion representing the first data portion associated with the first context type. The method includes accessing a second heterogeneous hash code including a third hash code portion representing a second data portion associated with the first context type. The method includes comparing the first heterogeneous hash code with the second heterogeneous hash code including determining similarity between the second hash code portion of the first heterogeneous hash code and the third hash code portion of the second heterogenous hash code.

Abstract: A pre-alarming method, a control method, and a control system for a harmful flow pattern in an oil and gas pipeline-riser system are provided. Support vector machines are trained. Through at least three pressure difference signals on the pipeline-riser system, an overall flow pattern in the pipeline-riser system is continuously and rapidly identified. Depending on monitoring on formation of a long liquid slug in a seabed pipeline and a quick response of the mean value of each pressure difference signal on a flow rate change, pre-alarming for a liquid slug caused by different mechanisms is realized, and liquid slug formation positions respectively of seabed pipeline and riser bottom are correspondingly pre-alarmed; after a pre-alarm is issued, there is enough time for a control device to respond, so as to avoid formation of the harmful flow pattern or damages caused by the harmful flow pattern.

Abstract: A method includes obtaining (i) device data associated with multiple electronic devices and (ii) advertisement data associated with one or more advertisement campaigns. The method also includes identifying first features and second features, where the first features correspond to the device data and the second features correspond to the advertisement data. The method further includes generating a graph relating usage history of the multiple electronic devices and one or more advertisement genres. The method also includes identifying a specified electronic device from among the multiple electronic devices and an advertisement segment from among the one or more advertisement campaigns using the first features, the second features, and the graph. In addition, the method includes providing the advertisement segment to the specified electronic device.

Abstract: In one embodiment, a computer-implemented method includes acquiring sequential user behavior data including one-dimensional data. The user behavior data is associated with a user. The method includes abstracting features from the sequential user behavior data to cover short-term and long-term timeframes. The method includes determining one or more properties of the user based on the features.

Abstract: An electronic device includes at least one processor configured to obtain user data associated with a plurality of devices from multiple data sources. The at least one processor is also configured to determine a static weight for each of the plurality of devices based on at least one source of the multiple data sources. The at least one processor is further configured to identify a portion of the plurality of devices that represents the plurality of devices based on the static weight and a dynamic weight. In addition, the at least one processor is configured to determine the dynamic weight for each of the portion of the plurality of devices while the portion of the plurality of devices is identified, where the dynamic weight is based on one or more sources of the multiple data sources.

Abstract: A strain of Yarrowia lipolytica has a deposit number of GDMCC No: 60782, and has been deposited on Sep. 20, 2019, in Guangdong Microbial Culture Collection Center located at Guangdong Institute of Microbiology, 5th Floor, Building 59, No. 100 Xianliezhong Road, Guangzhou City. A low-sugar and low-fat nutritional powder of desiccated coconut is prepared by: mixing a desiccated coconut with water, and heating; then pulping the heat-treated desiccated coconut, homogenizing an obtained coconut milk, adjusting a pH value, and sterilizing, to obtain a desiccated coconut suspension; inoculating a culture solution of the Yarrowia lipolytica in the desiccated coconut suspension after being cooled, and fermenting; and sterilizing an obtained fermentation broth, spray drying, and pulverizing, to obtain the low-sugar and low-fat nutritional powder of desiccated coconut.

Abstract: A pre-alarming method, a control method, and a control system for a harmful flow pattern in an oil and gas pipeline-riser system are provided. Support vector machines are trained. Through at least three pressure difference signals on the pipeline-riser system, an overall flow pattern in the pipeline-riser system is continuously and rapidly identified. Depending on monitoring on formation of a long liquid slug in a seabed pipeline and a quick response of the mean value of each pressure difference signal on a flow rate change, pre-alarming for a liquid slug caused by different mechanisms is realized, and liquid slug formation positions respectively of seabed pipeline and riser bottom are correspondingly pre-alarmed; after a pre-alarm is issued, there is enough time for a control device to respond, so as to avoid formation of the harmful flow pattern or damages caused by the harmful flow pattern.

Abstract: In one embodiment, a computer-implemented method includes acquiring sequential user behavior data including one-dimensional data. The user behavior data is associated with a user. The method includes abstracting features from the sequential user behavior data to cover short-term and long-term timeframes. The method includes determining one or more properties of the user based on the features.

Abstract: In one embodiment, a method includes receiving user behavior data and contextual information associated with the user behavior data, the contextual information including a first data portion associated with a first context type. The method includes generating, from the user behavior data and the contextual information using a hashing algorithm, a first heterogeneous hash code including a first portion representing the user behavior data and a second hash code portion representing the first data portion associated with the first context type. The method includes accessing a second heterogeneous hash code including a third hash code portion representing a second data portion associated with the first context type. The method includes comparing the first heterogeneous hash code with the second heterogeneous hash code including determining similarity between the second hash code portion of the first heterogeneous hash code and the third hash code portion of the second heterogenous hash code.

Abstract: Various techniques are provided to perform enhanced automatic speech recognition. For example, a subband analysis may be performed that transforms time-domain signals of multiple audio channels in subband signals. An adaptive configurable transformation may also be performed to produce single or multichannel-based features whose values are correlated to an Ideal Binary Mask (IBM). An unsupervised Gaussian Mixture Model (GMM) model fitting the distribution of the features and producing posterior probabilities may also be performed, and the posteriors may be combined to produce deep neural network (DNN) feature vectors. A DNN may be provided that predicts oracle spectral gains from the input feature vectors. Spectral processing may be performed to produce an estimate of the target source time-frequency magnitudes from the mixtures and the output of the DNN. Subband synthesis may be performed to transform signals back to time-domain.

Abstract: Various techniques are provided to perform enhanced automatic speech recognition. For example, a subband analysis may be performed that transforms time-domain signals of multiple audio channels in subband signals. An adaptive configurable transformation may also be performed to produce single or multichannel-based features whose values are correlated to an Ideal Binary Mask (IBM). An unsupervised Gaussian Mixture Model (GMM) model fitting the distribution of the features and producing posterior probabilities may also be performed, and the posteriors may be combined to produce deep neural network (DNN) feature vectors. A DNN may be provided that predicts oracle spectral gains from the input feature vectors. Spectral processing may be performed to produce an estimate of the target source time-frequency magnitudes from the mixtures and the output of the DNN. Subband synthesis may be performed to transform signals back to time-domain.

Abstract: A controller for a switched mode power supply (SMPS) is provided. The SMPS is equipped with a transformer having a primary side winding, a secondary winding, and an auxiliary winding. The controller includes a detection circuit for detecting a transition from a first output load condition to a second output load condition of the SMPS and a control circuit coupled to the detection circuit and being configured to output one or more control signals in response to the detected output load transition. Depending on the embodiment, the one or more control signals include a first control signal for turning on a power switch to cause a current flow in a primary winding of the SMPS and/or one or more second control signals for turning off one or more functional circuit blocks in the controller.