Abstract: A method of processing a query to a database from a query source is provided, comprising: receiving the query, the query in a first format supported by the query source; inputting the query into a first neural network; outputting, by the first neural network, the query in a second format, wherein the second format is a format supported by the database; receiving, from the database, a response to the query, the response in the second format; inputting the response to the query into a second neural network; outputting, by the second neural network, the response to the query in the first format; wherein each neural network is trained by inputting a first plurality of pairs of semi-structured data, each pair of semi-structured data comprising a sample query or response in the first format and the sample query or response in the second format.

Abstract: A system for prioritizing and resolving inconsistencies in digital evidence. The system includes a database containing a first type of data and a second type of data related to an incident record and an electronic computing device including an electronic processor. The electronic processor is configured to receive the first and second types of data from the database, determine an inconsistency between the first and second types of data, and determine an incident type from the incident record. The electronic processor is also configured to determine whether a priority of the determined inconsistency meets a threshold case impact level. When the priority of the inconsistency meets the threshold case impact level, the electronic processor is configured to take a first notification action and when the priority of the inconsistency does not meet the threshold case impact level, the electronic processor is configured to take a second notification action.

Abstract: Receive first context information (FCI) including entered in-field incident timeline information values from an in-field incident timeline application and a time associated with an entry of the FCI values. Access a mapping that maps in-field incident timeline information values to events having a pre-determined threshold confidence of occurring and identify an event associated with the received FCI. Determine a location associated with the entry of the FCI and a time period associated with the entry of the FCI. Access a camera location database and identify cameras that have a field of view including the location during the time period. Retrieve audio and/or video streams captured by the cameras during the time period. And provide the audio and/or video streams to machine learning training modules corresponding to machine learning models for detecting the event in and/or video streams for further training of the machine learning models.

Abstract: Obtaining potential match results for a reference image across a plurality of system sites is disclosed. A computing device of one of the system sites includes a signature generator identifiable as a first version amongst a plurality of versions of a respective plurality of possible signature generators. The computing device is configured to generate, within the first signature generator, a first signature corresponding to a cropped object portion of a larger image. The first signature is distinctive to the first version. The computing device is further configured to determine that the cropped object portion being processed within the computing device is a match result for a similar images search. A server is configured to receive the cropped object portion and the first signature from the computing device.

Abstract: A method of processing a query to a database from a query source is provided, comprising: receiving the query, the query in a first format supported by the query source; inputting the query into a first neural network; outputting, by the first neural network, the query in a second format, wherein the second format is a format supported by the database; receiving, from the database, a response to the query, the response in the second format; inputting the response to the query into a second neural network; outputting, by the second neural network, the response to the query in the first format; wherein each neural network is trained by inputting a first plurality of pairs of semi-structured data, each pair of semi-structured data comprising a sample query or response in the first format and the sample query or response in the second format.

Abstract: A media content item is evaluated for its “sentiment states.” That is, segments of the content item are determined to be, for example, “happy,” “exciting,” “sad,” “funny,” and the like. A “sentiment map” is created that delimits segments of the content item and contains the sentiment-state keywords associated with the segments. Some maps include an amplitude for each assigned sentiment keyword and a confidence value for the segment delimitation and for each keyword. As an exemplary use of the sentiment map, an advertisement broker matches the sentiments of his advertisement offerings with segments of a content item in order to place appropriate advertisements at times when they would be most favorably received. In another example, a recommender system recommends to a user a content item whose sentiment map compares favorably to that of a content item already enjoyed by the user.

Abstract: A system for prioritizing and resolving inconsistencies in digital evidence. The system includes a database containing a first type of data and a second type of data related to an incident record and an electronic computing device including an electronic processor. The electronic processor is configured to receive the first and second types of data from the database, determine an inconsistency between the first and second types of data, and determine an incident type from the incident record. The electronic processor is also configured to determine whether a priority of the determined inconsistency meets a threshold case impact level. When the priority of the inconsistency meets the threshold case impact level, the electronic processor is configured to take a first notification action and when the priority of the inconsistency does not meet the threshold case impact level, the electronic processor is configured to take a second notification action.

Abstract: A method for creating derivative advertisements is provided. The method includes identifying a visual element based on an analysis of a primary advertisement asset such that the visual element is representative of a branding message provided within the primary advertisement. The method also includes identifying a call-to-action (CTA) data element based on the analysis and creating a derivative advertisement asset having the visual element and CTA data element. A system and a non-transitory computer-readable storage medium containing software for executing the method are also provided.

Abstract: A system for prioritizing and resolving inconsistencies in digital evidence. The system includes a database containing a first type of data and a second type of data related to an incident record and an electronic computing device including an electronic processor. The electronic processor is configured to receive the first and second types of data from the database, determine an inconsistency between the first and second types of data, and determine an incident type from the incident record. The electronic processor is also configured to determine whether a priority of the determined inconsistency meets a threshold case impact level. When the priority of the inconsistency meets the threshold case impact level, the electronic processor is configured to take a first notification action and when the priority of the inconsistency does not meet the threshold case impact level, the electronic processor is configured to take a second notification action.

Abstract: A device, system and method for dispatching responders to patrol routes is provided. A device accesses: historical patrol route information indicating patrol routes taken for each of a plurality of time periods, each of the patrol routes taken by a respective uniquely identified responder; and historical crime information indicating actual crime rates for the patrol routes taken. The device receives a request to assign responders to respective patrol routes and responsively: determines available responders for the upcoming time period; determines, based on the available responders and the historical crime information, a mapping of the available responders to the respective patrol routes that reduces and/or minimizes an expected crime rate for the upcoming time period across the respective patrol routes; and based on the mapping, causes the available responders to be dispatched to the respective patrol routes for the upcoming time period.

Abstract: Techniques for enhancing strategic patrol planning and dispatch decision making based on gone on arrival prediction are provided. In one aspect, a crime prediction map may be retrieved. The crime prediction map may include incident locations and incident times, of predicted incidents occurring within a geographic area. The predictions may be based on historical data, the historical data including data from a computer aided dispatch (CAD) system. For each predicted incident location and incident time, a probability of gone on arrival (GOA) incident disposition may be calculated for a plurality of responder response times. The probability may be calculated based on the historical data from the CAD system.

Abstract: A first screen schedule for the linear delivery of video content is received and analyzed. The first screen schedule may include slots in which first screen advertisements could be displayed. Particular embodiments can then determine the feasibility of providing second screen advertisements based on the first screen schedule. Then, second screen advertisements that advertisers wish to display in conjunction with first screen advertisements are determined. When the linear delivery of the video content commences, particular embodiments monitor a messaging stream associated with the delivery of the video content. When an ad placement marker is encountered, an ad server may select a first screen advertisement. When the first screen advertisement is placed, the messaging stream identifies a first screen advertisement that is delivered. If the first screen advertisement has a corresponding second screen advertisement, particular embodiments output the second screen advertisement on a second screen environment.

Abstract: A media content item is evaluated for its “sentiment states.” That is, segments of the content item are determined to be, for example, “happy,” “exciting,” “sad,” “funny,” and the like. A “sentiment map” is created that delimits segments of the content item and contains the sentiment-state keywords associated with the segments. Some maps include an amplitude for each assigned sentiment keyword and a confidence value for the segment delimitation and for each keyword. As an exemplary use of the sentiment map, an advertisement broker matches the sentiments of his advertisement offerings with segments of a content item in order to place appropriate advertisements at times when they would be most favorably received. In another example, a recommender system recommends to a user a content item whose sentiment map compares favorably to that of a content item already enjoyed by the user.

Abstract: A system for prioritizing and resolving inconsistencies in digital evidence. The system includes a database containing a first type of data and a second type of data related to an incident record and an electronic computing device including an electronic processor. The electronic processor is configured to receive the first and second types of data from the database, determine an inconsistency between the first and second types of data, and determine an incident type from the incident record. The electronic processor is also configured to determine whether a priority of the determined inconsistency meets a threshold case impact level. When the priority of the inconsistency meets the threshold case impact level, the electronic processor is configured to take a first notification action and when the priority of the inconsistency does not meet the threshold case impact level, the electronic processor is configured to take a second notification action.

Abstract: A device, system and method for dispatching responders to patrol routes is provided. A device accesses: historical patrol route information indicating patrol routes taken for each of a plurality of time periods, each of the patrol routes taken by a respective uniquely identified responder; and historical crime information indicating actual crime rates for the patrol routes taken. The device receives a request to assign responders to respective patrol routes and responsively: determines available responders for the upcoming time period; determines, based on the available responders and the historical crime information, a mapping of the available responders to the respective patrol routes that reduces and/or minimizes an expected crime rate for the upcoming time period across the respective patrol routes; and based on the mapping, causes the available responders to be dispatched to the respective patrol routes for the upcoming time period.

Abstract: An advertisement system includes an advertising manager that receives a content-event indicator, which indicates playback of an advertisement for viewing on a first display screen at a media-playback device or indicates playback of recorded content that includes advertisements. The advertising manager can determine an optimal time offset as a duration of time before or after playback of an advertisement to the start of an event that is associated with the advertisement for viewing on a second display screen at a mobile device. For recorded content, an optimization schedule is determined that replaces and time-shifts advertisements during playback of the recorded content. The advertising manager can also determine a fulfillment criterion for a product or service based on latency constraints to indicate a duration of time within which fulfillment of the product or service is expected when offered as a second advertisement corresponding to a first advertisement.

Abstract: A media content item is evaluated for its “sentiment states.” That is, segments of the content item are determined to be, for example, “happy,” “exciting,” “sad,” “funny,” and the like. A “sentiment map” is created that delimits segments of the content item and contains the sentiment-state keywords associated with the segments. Some maps include an amplitude for each assigned sentiment keyword and a confidence value for the segment delimitation and for each keyword. As an exemplary use of the sentiment map, an advertisement broker matches the sentiments of his advertisement offerings with segments of a content item in order to place appropriate advertisements at times when they would be most favorably received. In another example, a recommender system recommends to a user a content item whose sentiment map compares favorably to that of a content item already enjoyed by the user.

Abstract: Particular embodiments provide a system to support orchestration of media assets that are delivered across multiple media spaces, time, and multiple device types in a coordinated manner. The system exposes a sequence of media assets to a user under media space, time, and device constraints. A campaign may be broken up into multiple media assets (e.g., different creatives, such as advertisement creatives). To target a user across different media spaces, the system may generate spanning rules based on metadata concerning multiple media spaces, time, and multiple device types for the campaign. An analytics service may provide analytics information for the delivered media assets to allow the coordination to continue in the sequence. Also, the analytics information may be used to change the spanning rules for future delivery of media assets.

Abstract: Particular embodiments take first screen media assets and can derive second screen appropriate media assets. Then, particular embodiments may determine a target application for the second screen media assets. Layout characteristics for the target application is determined. Particular embodiments correlate the layout characteristics and the second screen media assets to select a second screen media asset for the target application. The correlation may generate scores based on the second screen media assets and the layout characteristics to determine a best fit of a second screen media asset for the target application. Then, the second screen media asset may be transcoded based on the characteristics of the target application or target device. After transcoding, particular embodiments may create a view of the second screen media asset in the target application, such as an overlay of the second screen media asset over a screen capture of the target application.

Abstract: Disclosed are methods and apparatus for triggering (i.e., initiating, launching, or running) a software application (e.g., an interactive game application) or other computer program or application on a user device (e.g., a computer, etc.) based on multimedia content being provided to (e.g., displayed on) a different user device (e.g., a different computer, a television, etc.). The methods may comprise displaying, using a first device, multimedia content. A multimedia-analysis process may then be performed to determine an attribute of the multimedia content being displayed. These attributes may be used to acquire (e.g., select or produce) a software application comprising one or more elements that relate to the multimedia content. The software application may then be launched (i.e., triggered, initiated, or run) on a second device, the second device being different from the first device.