Abstract:
More efficient mobile device location data can be obtained by estimating a most likely location point in a coverage pattern using a kernel density estimation technique. The kernel density estimation technique can provide a continuous estimate of the most frequented locations of a mobile device(s) within a coverage area. For each wireless sector, the collected location data can grouped to the closest geographic coordinate system, and an inference can be made based on the grouped data.
Abstract:
More efficient mobile device location data can be obtained by estimating a most likely location point in a coverage pattern using a kernel density estimation technique. The kernel density estimation technique can provide a continuous estimate of the most frequented locations of a mobile device(s) within a coverage area. For each wireless sector, the collected location data can grouped to the closest geographic coordinate system, and an inference can be made based on the grouped data.
Abstract:
Aspects of the subject disclosure may include, for example, receiving registration information indicating a plurality of registered communication devices in proximity to a set-top box, the plurality of registered communication devices being registered according to a communication exchange between the set-top box and each of the plurality of registered communication devices, identifying a group of users associated with the plurality of registered communication devices, and determining characteristics for the group of users according to the identifying of the group of users, correlating the characteristics for the group of users, selecting a directed advertisement according to the correlating of the characteristics, and providing the directed advertisement to the set-top box for presentation. Other embodiments are disclosed.
Abstract:
Methods, computer-readable media and apparatuses for predicting an amount of network infrastructure needed for a new neighborhood are disclosed. A processor generates a plurality of different user profiles based upon demographic data of existing customers, historical utilization data and historical usage data, determines a demographic of a new neighborhood, correlates one of the plurality of different user profiles to the new neighborhood based upon the demographic of the new neighborhood and predicts the amount of network infrastructure to be deployed in the new neighborhood based upon the one of the plurality of different user profiles that is correlated to the demographic of the new neighborhood.
Abstract:
Determining referenced based location information for a wireless radio network is described. Referenced based location information can include determining location reference information and corresponding location offset information based on location information. In an aspect, location information can be timed fingerprint location information. Location offset information can be communicated in a wireless network at a lower operational cost than the associated location information. As such, use of referenced based location information for a wireless network can reduce bandwidth consumption as compared to location information communicated at similar intervals. This is particularly true in large wireless networks. Moreover, the use of referenced based location information for determining timed fingerprint location information can be highly attractive in light of timed fingerprint location information facilitating location information for many non-GPS enabled devices and being associated with significant increases in the frequency and density of location event requests.
Abstract:
Concepts and technologies are disclosed herein for providing and/or interacting with a profile verification service. A processor executing a profile verification service can receive a request to verify a user profile associated with a user of a social networking application. The processor can identify a computing device associated with the user profile, obtain location data that relates to the user profile and the computing device, and identify an activity associated with the computing device based upon the location data. The processor can determine if the user profile is accurate based upon the activity identified. If a determination is made that the user profile is accurate, the processor can verify the user profile. If a determination is made that the user profile is not accurate, the processor can update the user profile.
Abstract:
To provide mobile device coupons and discounts, a mobile device couponing system is provided to generate and send coupons to mobile devices based on a probability that the mobile device is within a defined region. The value of the coupons can be inversely proportional to the probability that the mobile device is within the defined region. The probabilities can be precomputed by calculating the probability that each cell in grid in or around the defined region is in the defined region using a set of standardized location resolutions. When location information is received from the mobile device, the probability that the mobile device is within the defined region can be retrieved based on the closest cell that matches the location indicated in the location information and the standardized location resolution closest to the location resolution indicated by the mobile device.
Abstract:
Determining levels of geographic redundancy among radios of a wireless radio network is disclosed. The level of geographic redundancy for a radio can affect the determination of location information for a user equipment (UE) on the wireless radio network. The disclosed subject matter can be employed in conjunction with timed fingerprint location (TFL) technologies to facilitate selection of radios employed in determining time values for TFL location determination. Levels of geographic redundancy can be employed to rank or order radios of a wireless radio network so as to reduce the likelihood of using geographically redundant radios in location determination. Further, rules can be selected to adjust threshold values and equations employed in determining the levels of geographic redundancy. Moreover, rules can be selected to apply boundary conditions to reduce the number of determinations formed for a set of radios of the wireless radio network.
Abstract:
A method performed by a processing system including at least one processor includes defining a proposal for a proposed machine learning model, identifying an existing machine learning model, where the existing machine learning model shares a similarity with the proposed machine learning model, evaluating a fitness of the existing machine learning model for reuse in building the proposed machine learning model, building a new machine learning model that is consistent with the proposal for the proposed machine learning model by reusing a portion of the existing machine learning model, and monitoring a performance of the new machine learning model in a deployment environment.
Abstract:
A sorter receives a list of elements to be sorted. An element of the list is supplied to a selected one of a plurality of processing units to be processed. The selected one of the processing units sends the element to one of a plurality of list element cells, which rank orders the elements among other elements in the same list element storage as well as storing the position of each element from the original list. Each of the plurality of list element cells processes and stores a different range of element values. The element being processed is stored in sorted order in the list element cell that has an element value range that encompasses the value of the element of the list.