Abstract: In some implementations, a device may obtain a set of one or more communication events associated with a user device, where the set of one or more communication events indicates respective values of one or more communication parameters. The device may determine, for each communication event, whether one or more rules are applicable to that communication event based on a comparison of values of the one or more communication parameters of that communication event and values of the one or more communication parameters of a previous communication event to that communication event in time. The device may assign, for each communication event, a movement state for that communication event based on whether a score for that communication event satisfies a threshold, where the score is a summation of one or more weights of respective rules applicable to that communication event.

Abstract: In some implementations, a device may obtain location data associated with one or more user devices, wherein the location data indicates, for the one or more user devices, respective geographic locations and respective durations. The device may determine, based on the location data, one or more point of interest (POI) locations associated with respective user devices from the one or more user devices. The device may determine, for each user device from the one or more user devices, probability values for respective POI locations from the one or more POI locations, wherein the probability values indicate likelihoods of user visits at the respective POI locations. The device may determine one or more probability metrics for the respective POI locations, from the one or more POI locations, associated with user visits to the respective POI locations. The device may store the one or more probability metrics.

Abstract: A device may receive data identifying characteristics, location dependent characteristics, and a network service associated with target consumers, such as households or businesses, and may process the data, with one or more propensity models, to determine propensities of the target consumers to utilize the network service. The device may process the data and the propensities, with a state transition model, to calculate probabilities that the target households will utilize the network service, and may determine utilization states of the target consumers over time based on the probabilities that the target consumers will utilize the network service. The device may aggregate the utilization states of the target consumers to determine penetration rates for the network service, and may perform one or more actions based on the penetration rates for the network service.

Abstract: Disclosed are systems and methods for a computerized, network-based framework that provides enhanced algorithmic mechanisms for improving GSM location determinations for cell tower communications. The disclosed framework's operations can provide improved visit determinations/estimates and increased value for a user's wireless/cellular data, whereby optimized parameters from and/or of the signals associated with a user device (e.g., mobile device) and tower radio communications therebetween can be leveraged to discern the POIs a user has visited and/or is currently visiting. The location tracking of the user, within a predisposed uncertainty sector correlated among the user's device communications with corresponding radio towers, provides network coverage to track specific locations among POIs for which the user has visited and/or is currently located.

Abstract: In some implementations, the techniques described herein relate to a method including: receiving a set of communication events from a user equipment to a base station; obtaining a set of geocoordinates corresponding to the user equipment; calculating an angular mean for the set of geocoordinates; sorting the set of geocoordinates based on distances between the set of geocoordinates and a location of the base station to generate an ordered list of geocoordinates; calculating a k-th order statistic of the ordered list of geocoordinates; generating a predicted location based on the angular mean and the k-th order statistic; and assigning the predicted location as a current location of the user equipment in a cellular core network.

Abstract: One or more systems and/or methods are provided for identifying one or more target roads of a point of interest (POI) and/or identifying impressions of the POI. In an example, a first target road of a POI may be identified, wherein the POI is viewable from the first target road. A plurality of locations of a User Equipment (UE) may be determined. A set of locations, of the plurality of locations, that are within a threshold distance of a road object corresponding to the first target road may be identified. It may be determined that the UE is associated with the impression based upon the set of locations meeting a threshold amount of locations.

Abstract: The present teaching relates to impression count determination. A forecasting time series (TS) model is established based on measured impression counts (MI-counts). Metrics are calculated from MI-counts from a sub-range of a profile. Different types of data characteristics are detected based on the metrics. Hybrid correction applied to the profile is determined based on the detected data characteristics. Corrected impression counts (I-counts) for the profile are generated via the hybrid correction operation based on the MI-counts and I-counts estimated from the forecasting TS models and provided for determining a level of viewership of the content at the site.

Abstract: In some implementations, a device may obtain a set of one or more communication events associated with a user device, where the set of one or more communication events indicates respective values of one or more communication parameters. The device may determine, for each communication event, whether one or more rules are applicable to that communication event based on a comparison of values of the one or more communication parameters of that communication event and values of the one or more communication parameters of a previous communication event to that communication event in time. The device may assign, for each communication event, a movement state for that communication event based on whether a score for that communication event satisfies a threshold, where the score is a summation of one or more weights of respective rules applicable to that communication event.

Abstract: In some implementations, a device may obtain location data associated with one or more user devices, wherein the location data indicates, for the one or more user devices, respective geographic locations and respective durations. The device may determine, based on the location data, one or more point of interest (POI) locations associated with respective user devices from the one or more user devices. The device may determine, for each user device from the one or more user devices, probability values for respective POI locations from the one or more POI locations, wherein the probability values indicate likelihoods of user visits at the respective POI locations. The device may determine one or more probability metrics for the respective POI locations, from the one or more POI locations, associated with user visits to the respective POI locations. The device may store the one or more probability metrics.

Abstract: The present teaching relates to impression count determination. A forecasting time series (TS) model is established based on measured impression counts (MI-counts). Metrics are calculated from MI-counts from a sub-range of a profile. Different types of data characteristics are detected based on the metrics. Hybrid correction applied to the profile is determined based on the detected data characteristics. Corrected impression counts (I-counts) for the profile are generated via the hybrid correction operation based on the MI-counts and I-counts estimated from the forecasting TS models and provided for determining a level of viewership of the content at the site.

Abstract: A neural network is used to model to model the joint distribution of attributes across multiple health surveys. These multiple health surveys include large scale survey datasets and small scale survey datasets. The neural network model is trained using a combined dataset of the large scale survey datasets and the small scale survey datasets. The large scale survey datasets and the small scale survey datasets may include missing value indicators. The joint distribution of attributes modeled by the neural network model are the used to impute substitute values for the missing values to thereby create an output large scale dataset that does not include missing values.

Abstract: A method, non-transitory computer readable medium and apparatus for calculating a by spot occupancy of a parking lot are disclosed. For example, the method includes receiving an indication of a triggering event, sending a query to receive a first image and a second image in response to the triggering event, receiving the first image and the second image, analyzing the first image and the second image to determine a change in an occupancy status of a parking spot within the parking lot and calculating the by spot occupancy of the parking lot based on the change in the occupancy status of the parking spot.

Abstract: A neural network is used to model to model the joint distribution of attributes across multiple health surveys. These multiple health surveys include large scale survey datasets and small scale survey datasets. The neural network model is trained using a combined dataset of the large scale survey datasets and the small scale survey datasets. The large scale survey datasets and the small scale survey datasets may include missing value indicators. The joint distribution of attributes modeled by the neural network model are the used to impute substitute values for the missing values to thereby create an output large scale dataset that does not include missing values.

Abstract: A neural network is used to model to model the joint distribution of attributes across multiple health surveys. These multiple health surveys include large scale survey datasets and small scale survey datasets. The neural network model is trained using a combined dataset of the large scale survey datasets and the small scale survey datasets. The large scale survey datasets and the small scale survey datasets may include missing value indicators. The joint distribution of attributes modeled by the neural network model are the used to impute substitute values for the missing values to thereby create an output large scale dataset that does not include missing values.

Abstract: A method, system, and apparatus for license plate relighting comprises collecting an image of a license plate, performing license plate recognition on the image of the license plate; calculating a confidence metric for the license plate recognition; and performing a shadow detection and relighting method if the confidence metric is below a predetermined threshold, comprising identifying a shaded region of said license plate, determining if the shaded region is actually shaded, and relighting the actually shaded region.

Abstract: Methods and systems for localizing numbers and characters in captured images. A side image of a vehicle captured by one or more cameras can be preprocessed to determine a region of interest. A confidence value of series of windows within regions of interest of different sizes and aspect ratios containing a structure of interest can be calculated. Highest confidence candidate regions can then be identified with respect to the regions of interest and at least one region adjacent to the highest confidence candidate regions. An OCR operation can then be performed in the adjacent region. An identifier can then be returned from the adjacent region in order to localize numbers and characters in the side image of the vehicle.

Abstract: A method, system, and apparatus for license plate relighting comprises collecting an image of a license plate, performing license plate recognition on the image of the license plate; calculating a confidence metric for the license plate recognition; and performing a shadow detection and relighting method if the confidence metric is below a predetermined threshold, comprising identifying a shaded region of said license plate, determining if the shaded region is actually shaded, and relighting the actually shaded region.

Abstract: A method, system, and apparatus for parking occupancy detection comprises collecting video of a blockface with at least one video recording module, identifying a number of possible parking spaces along the blockface in the collected video, defining, a region of interest for each of the possible parking spaces, detecting a time dependent occupancy of the defined regions of interest for each of the possible parking spaces, and reporting the time dependent occupancy. Drift correction of the recorded video and ground truth comparisons of occupancy determinations may be provided.

Abstract: A method, system, and apparatus for parking occupancy detection comprises collecting video of a blockface with at least one video recording module, identifying a number of possible parking spaces along the blockface in the collected video, defining, a region of interest for each of the possible parking spaces, detecting a time dependent occupancy of the defined regions of interest for each of the possible parking spaces, and reporting the time dependent occupancy. Drift correction of the recorded video and ground truth comparisons of occupancy determinations may be provided.

Abstract: A method, system, and apparatus for video frame alignment comprises collecting video data comprising at least two video frames; extracting a line profile along at least one line profile in each of the at least two video frames; selecting one of the at least two video frames as a reference video frame; segmenting each of the at least one line profiles into a plurality of segmented line profile group segments; aligning the plurality of segmented line profiles with the segmented line profiles in the reference video frame; translating each of the at least two video frames for each of the plurality of corresponding segmented line profile alignments; and removing a camera shift from the at least two video frames according to the translation and alignment of the plurality of segmented line profiles with the plurality of segmented line profile in the reference video frame.