Back to results
US 10,949,880 B2Granted
Method for optimizing media and marketing content using cross-platform video intelligence
AssigneeICX Media, Inc.
InventorsJyotika Singh, Rebecca Bilbro, Michael Avon, Scott Bowen, Dan Jolicoeur, Serge Matta
Life of this patent
Estimating the term…
Abstract
The invention is directed to a computer-implemented method of analyzing video interactions on internet-supported computer platforms, such as online social media platforms, to extract video intelligence, i.e. unique insights and recommendations for audience engagement optimization, network growth, advertising, and marketing purposes.
Claims25
A computer-implemented method to generate predictive models comprising the following steps:
- a. using textual metadata associated with one or more social media video files to extract one or more features from audio, video, and/or text content;
- b. transforming the one or more extracted features into one or more numeric feature vectors representing the one or more social media video files in multidimensional vector space;
- c. using the one or more numeric feature vectors to classify social media video content, content consumers, and/or content producers into demographics;
- d. providing a model training interface capable of re-training when new data becomes available, automated data cleaning, a single, unified computer-implemented interface for social media language classification including computer-implemented methods for classification of arbitrary text that is social media platform agnostic, wherein the single, unified computer-implemented interface is capable of allowing for input of raw text input and for output of predictions over one or more machine learning classifiers;
- and
- e. sharing functionality across the single, unified computer-implemented interface, including text vectorization and normalization methodologies for automated cleaning and processing raw social media text;
- f. generating one or more analytics and/or recommendations to the content producers or content consumers, wherein the one or more analytics and/or recommendations comprise computer-generated information relating to video publishing, sourcing, targeted advertising, or combinations thereof, wherein the computer-generated information relating to video publishing, sourcing, targeted advertising, or combinations thereof are based on patterns and preferences of the content consumers and the content producers, including demographics, topical content, themes, optimal post time, viewing frequency, social media platform being used, peer sets, viewing length, messaging, marketing information comprising title, description, keywords, hashtags, or combinations thereof;
- g. wherein the one or more extracted features are transformed to numeric feature vectors by creating a set of words and textual features appearing in comments and descriptions or in metadata associated with the one or more social media video files;
- f. wherein the one or more extracted features comprise names, comments, typographic patterns, emojis, hyperlink usage, posting behavior, or combinations thereof, and wherein the one or more extracted features are extracted from text content using natural language processing.
6 more elements
A computer-implemented method using one or more computer models using natural language processing, feature extraction, numeric feature vectorization, and text classifiers to extrapolate detailed consumer profiles from social media video content consumption and interaction patterns, including the following steps:
- a. automatically creating content consumer profiles using combined results of one or more analytics and/or recommendations comprising computer-generated information relating to video publishing, sourcing, targeted advertising, or combinations thereof, wherein the computer-generated information relating to video publishing, sourcing, targeted advertising, or combinations thereof are based on patterns and preferences of video consumers, including demographics, topical content, themes, optimal post time, viewing frequency, social media platform being used, peer sets, viewing length, messaging, marketing information comprising title, description, keywords, hashtags, or combinations thereof;
- b. automatically establishing relationships between the content consumer profiles and video content consumption behaviors and preferences;
- c. automatically identifying special interest groups amongst video consumers;
- f. detecting social media content related to a subject, topic, person, TV show and/or movie title based on automatic similar search term generation, text vectors, and content classifications;
- d. detecting social media content related to a subject, topic, person, TV show and/or movie title based on automatic similar search term generation, text vectors, and content classifications;
- e. identifying consumers interested in the detected social media content based on consumption patterns, peer analysis and look-alike modeling;
- and
- f. combining the consumers or the detected social media content with other predictions on consumer demographics, location, and content preferences to create audience segments that are capable of being used for content targeting, ad targeting, audience analytics, recommendations for content providers, or combinations thereof;
- g. making recommendations to a content provider on video content, posting times, video publishing, sourcing, or targeted advertising;
- h. wherein the demographics comprise age, gender and ethnicity, location, direct marketing area, brand or individual creator status, low-level categorical and topical attributes, high-level categorical and topical attributes, political views, religious views, cross-platform account linkages, education level and income range, or combinations thereof.
7 more elements
Dependencies are inferred from the claim text.