Integral Ad Science, Inc.

Methods, systems, and media for providing contextual information associated with webpages are provided. In some embodiments, the method comprises: receiving a plurality of risk tolerance values from a first advertiser; accessing a first webpage through a first universal resource locator (URL), wherein the first webpage contains at least one dynamic advertising region; determining, using a machine learning model, (i) that the first URL is a cybersquatting attempt of a second URL based on a comparison of a first domain name associated with the first URL and a second domain name associated with the second URL; (ii) a first plurality of sentiments associated with content items within the first webpage, a plurality of keywords associated with the first webpage, and a similarity score between the first plurality of sentiments and the plurality of keywords; and (iii) a plurality of sentiment risk scores, wherein a sentiment risk score for at least one sentiment in the first plurality of sentiment risk scores is based on searching an approval list using at least one of the first plurality of sentiments as a search query; based on determining that the first URL is the cybersquatting attempt of the second URL, modifying an aggregate risk score by a first value from the plurality of risk tolerance values; based on determining that the similarity score is below a similarity threshold, modifying the aggregate risk score by a second value from the plurality of risk tolerance values; based on the plurality of sentiment risk scores, modifying the aggregate risk score by a third value from the plurality of risk tolerance values; determining that the aggregate risk score is within a first range of predetermined values; and in response to determining that the aggregate risk score is within the first predetermined range of values, associating at least one of the first webpage or the first domain name associated with the first URL with an exclusion list associated with the first advertiser.

Methods, systems, and media for streaming a media content item containing advertising content are provided. In some embodiments, the method comprises: identifying, in a manifest file at a server used to stream media content items, an indication that an advertising content item is to be played back at a first playback position for a first duration when streaming a media content item; determining a retrieval time to retrieve the advertising content item from an advertising source for an advertisement break; inserting, into the manifest file at the server used to stream media content items, a retrieval indication that the advertising content item is to be retrieved from the advertising source at a second playback time; retrieving the advertising content item from the advertising source when streaming of the media content item has reached the second playback position; determining a remaining amount of time corresponding to an unfilled portion of the advertisement break based on a duration of the advertising content item; based on the remaining amount of time corresponding to the unfilled portion of the advertising break, retrieving at least one additional advertisement content item for insertion in the unfilled portion of the advertising break, where the at least one additional advertisement content item is selected from a plurality of processed advertising content items received from a plurality of network sources, where each of the plurality of network sources has uploaded and assigned at least a first advertising content item having a first duration and at least a second advertising content item having a second duration, and where the first duration associated with the first advertising content item is determined to be within a first range and the second duration associated with the second advertising content item is determined to be within a second range; and collapsing the remaining portion of the advertising break to the end of the at least one additional advertising content item that was retrieved.

In accordance with some embodiments of the disclosed subject matter, methods, systems, and media for setting and using an advertisement frequency cap based on causal conversions or impact of advertisements are provided.

Methods, systems, and media for review content traffic are provided. In accordance with some embodiments of the present invention, a method, implemented on a processor, for reviewing advertisement placement on webpages is provided. The method can include: receiving a plurality of uniform resource locators (URLs) relating to the placement of a plurality of advertisements associated with an advertiser; obtaining, for the plurality of URLs, a plurality of images and a plurality of advertisement traffic and rating information relating to each of the plurality of URLs, wherein each of the plurality of images shows the placement of an advertisement on the page accessed by a URL; presenting a traffic display that includes the plurality of images for the plurality of pages, wherein the plurality of images are user selectable; and, in response to selecting one of the plurality of images, presenting the advertiser with the advertisement traffic and rating information for a URL corresponding to the selected image and providing the advertiser with an opportunity to provide feedback information for the rating information associated with the URL.

Mechanisms for inhibiting the transmission of media content based on frequency and exposure measurements are provided. In some embodiments, the method comprises: receiving consumption information of a plurality of content items associated with a plurality of viewers; determining, from the consumption information, exposure frequency for each viewer of the plurality of viewers, exposure time of a set of content items provided to each viewer of the plurality of viewers, and lift for each viewer of the plurality of viewers, wherein the lift corresponds to a difference of an expected number of treated conversions and an expected number of untreated conversions over a sum of the expected number of treated conversions and the expected number of untreated conversions; determining an optimal exposure frequency and an optimal exposure time based on the exposure frequency, the exposure time, and the lift; generating (i) a first user segment in which a plurality of additional content items should be provided to a viewer of the plurality of viewers, (ii) a second user segment in which at least one additional content item should be provided to the viewer of the plurality of viewers, and (iii) a third user segment in which no additional content items should be provided to the viewer of the plurality of viewers; receiving a request for a content item from a user device; determining whether the user device should be assigned with the first user segment, the second user segment, or the third user segment; in response to determining that the user device should be assigned to the first user segment, transmitting the plurality of additional content items until determining that the user device should be assigned to the second user segment based on updated exposure frequency information and updated exposure time information; continuing to transmit the at least one additional content item to the user device while the user device is positioned within the second user segment until determining that the user device should be assigned to the third user segment based on the updated exposure frequency information and the updated exposure time information; and inhibiting the presentation of any additional content items on the user device in response to the user device being positioned within the third user segment.

Methods, systems, and media for providing automated review of incoming messages in a group messaging service are provided. In some embodiments, the method includes: receiving, at a user device from a server, an incoming message to a particular message group in a group messaging service, wherein the user device is associated with the particular message group; determining, using a machine learning classifier at the user device, (i) a context of the particular message group based on a plurality of content items previously submitted by a plurality of users to the particular message group; (ii) a plurality of sentiments in the incoming message; (iii) a plurality of similarity scores, wherein each similarity score is a probability that a sentiment from the plurality of sentiments is similar to the context of the particular message group; and (iv) an aggregate similarity score based on the plurality of similarity scores is within a first range of predetermined values; and, in response to determining the aggregate similarity score is within the first range of predetermined values, displaying a preview of the incoming message on the user device, wherein the preview includes at least one sentiment from the plurality of sentiments.

Methods, systems, and media for generating one or more custom models, such as artificial intelligence models or machine learning models, using a multimedia understanding model. More particularly, the multimedia understanding model can be a large foundational model that is trained using image data, video data, audio data, text data, and/or page data extracted from multiple content items, where the multimedia understanding model can generate, for a given content item, a unified embedding for use with one or more machine learning models (e.g., a classification server executing a classification model that classifies the content of the content item, such as a video content item, into each of twelve defined risk categories) and/or applications (e.g., an application that generates groups of content items that represent daily trends, a search engine application that provides matching content items based on text inputs, image inputs, audio inputs, video inputs, etc., a classification application that generates new or additional categories for classifying the content of a content item, etc.).

Systems, methods, and media for correlating information corresponding to multiple related frames on a web page are provided. In some embodiments, a method for identifying corresponding iframes on a web page to a server is provided, the method comprising: loading a web page including a plurality of iframes; receiving the requested content and identifying information for a first iframe; loading an event listener; determining a location of the first iframe in the structure of the web page; transmitting a message to a second iframe including the first identifying information and location of the first iframe; receiving, using the event listener, a message from the second iframe including second identifying information and location of the second iframe; determining that the second iframe is nested within the first iframe; and transmitting a third message to a server including at least the second identifying information and location of the second iframe.

Methods, systems, and media for enhancing a blind URL escrow with real-time bidding exchanges are provided. In some embodiments, the method comprises: receiving, using a hardware processor, a URL associated with an advertisement placement from a seller; generating a rating for the URL received from the seller; determining metadata based on the URL and a domain name associated with the seller; generating an anonymous label corresponding to the URL that includes the generated rating and the determined metadata; transmitting the anonymous label to a real-time bidding exchange; and causing a verification beacon to be inserted into an advertisement to be displayed in the advertisement placement in response to a buyer purchasing an opportunity to place an advertisement in the advertisement placement using the anonymous label through the real-time bidding exchange.

Systems, methods, and media for detecting suspicious activity in connection with advertisement impressions are provided. In some embodiments, the method includes: collecting advertisement impression information associated with a plurality of pages; determining, from the collected advertisement impression information, an indication of whether a browser application detected that an advertisement displayed on a webpage was viewable in a browser window; determining, from the collected advertisement impression information, a plurality of viewability statistics for each of the plurality of pages, wherein each viewability statistic indicates a likelihood of whether an advertisement displayed on a webpage was viewable in a browser window; comparing the plurality of viewability statistics with the indication from the browser application; determining a viewability score for the advertisement impression based on the comparison; and identifying the advertisement impression as likely to be suspicious based on the determined viewability score.