# US 12,118,047 B2 · Systems and methods for multi-dimensional electronic content performance predictor and decision automation > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,118,047 B2 - Document ID: US12118047B2 - Status: Granted - Kind: B2 - Application number: 15/955,410 - Filed: 2018-04-17 - Published: 2023-06-15 - Granted: 2024-10-15 - Earliest claimed: 2018-04-17 - Assignees: Yahoo Assets LLC - Applicants: Yahoo Ad Tech LLC - Inventors: Raj Yadav; Abhiraj Arumughan Koyamkattil; Ankit Katiyar; Tushar Mahule; Aileen Barrameda - Web page: https://patentgrep.com/patent/US12118047B2 ## CPC - G06F 16/9535: Search customisation based on user profiles and personalisation - G06F 16/24578: {using ranking} - G06N 3/0499: Feedforward networks - G06N 3/084: Backpropagation, e.g. using gradient descent - G06N 20/00: Machine learning - G06N 3/09: Supervised learning - H04L 67/1097: for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS] - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract Systems and methods are disclosed for vertical integration of a multi-dimensional machine learning algorithm with a digital content management system. A variety of machine learning algorithms are implemented to aid publishers in the discovery of digital content. In contrast to prior methods, which perform retroactive assessments on digital content performance, the instant methods train machine learning algorithms to predict future performance of digital content items. As digital content becomes available to publishers, machine learning algorithms analyze the digital content items in order to present editorial actions publishers should take via a digital content management dashboard. ## Claims ### Claim 1 A computer-implemented method for generating optimized digital content for electronic display, the method comprising: receiving, at a prediction engine of a server, one or more digital content items from one or more content partners, the one or more digital content items including stored data; analyzing, using a machine learning algorithm and a neural network, each of the one or more digital content items against performance criteria to determine a corresponding potential total traffic value; comparing, by the prediction engine, the corresponding potential total traffic value for each of the one or more digital content items against a traffic threshold; based on comparing the corresponding potential total traffic value for each of the one or more digital content items against the traffic threshold, determining, by the prediction engine, a first subset of the one or more digital content items and a second subset of the one or more digital content items, wherein the first subset of the one or more digital content items include the one or more digital content items that have met or exceeded the traffic threshold, and the second subset of the one or more digital content items include the one or more digital content items that have not met the traffic threshold; transmitting, by a decision engine of the server, the first subset of the one or more digital content items that have met or exceeded the traffic threshold to a digital content management system; generating, by the decision engine of the server, a ranking of the first subset of the one or more digital content items and a corresponding action recommendation determined by the machine learning algorithm and the neural network; and conveying to a user, by the decision engine of the server, the ranking of the first subset of the one or more digital content items and the corresponding action recommendation as action buttons, the action buttons including a first button that instantly publishes a top emerging digital content item to mobile devices and a second button that provides insight information comprising expected performance. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein training data of the neural network is trained over a predetermined period of time. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein training data of the neural network is stored in a distributed file system. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein the digital content items that exceed the traffic threshold, exceed a predetermined web site traffic value. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, wherein the machine learning algorithm utilizes the neural network to forecast digital content item performance. ### Claim 6 (depends on 5) The computer-implemented method of claim 5, wherein the machine learning algorithm utilizes supervised learning, unsupervised learning, reinforcement learning and/or semi-supervised learning; and wherein the neural network is multilayered and includes an input layer and one or more of a hidden layer and an output layer. ### Claim 7 (depends on 3) The computer-implemented method of claim 3, wherein the one or more action recommendations received from the digital content management system is fed back into the neural network to further train the stored training data. ### Claim 8 A computer system for executing a machine learning algorithm for generating optimized digital content, comprising: a memory having processor-readable instructions stored therein; and a processor configured to access the memory and execute the processor-readable instructions, which when executed by the processor configures the processor to perform plurality of functions, including functions for: receiving, at a prediction engine of a server, one or more digital content items from one or more content partners, the one or more digital content items including performance information; analyzing, using a machine learning algorithm and a neural network, each of the one or more digital content items against performance criteria to determine a corresponding potential total traffic value; comparing, by the prediction engine, the corresponding potential total traffic value for each of the one or more digital content items against a traffic threshold; based on comparing the corresponding potential total traffic value for each of the one or more digital content items against the traffic threshold, determining, by the prediction engine, a first subset of the one or more digital content items and a second subset of the one or more digital content items, wherein the first subset of the one or more digital content items include the one or more digital content items that have met or exceeded the traffic threshold, and the second subset of the one or more digital content items include the one or more digital content items that have not met the traffic threshold; transmitting, by a decision engine of the server, the first subset of the one or more digital content items that have met or exceeded the traffic threshold to a digital content management system; generating, by the decision engine of the server, a ranking of the first subset of the one or more digital content items and a corresponding action recommendation determined by the machine learning algorithm and the neural network; and conveying to a user, by the decision engine of the server, the ranking of the first subset of the one or more digital content items and the corresponding action recommendation as action buttons, the action buttons including a first button that instantly publishes a top emerging digital content item to mobile devices and a second button that provides insight information comprising expected performance. ### Claim 9 (depends on 8) The computer system of claim 8, wherein training data of the neural network is trained over a predetermined period of time. ### Claim 10 (depends on 8) The computer system of claim 8, wherein training data of the neural network is stored in a distributed file system. ### Claim 11 (depends on 8) The computer system of claim 8, wherein the machine learning algorithm utilizes the neural network to forecast digital content item performance. ### Claim 12 (depends on 8) The computer system of claim 8, wherein the digital content items that exceed the traffic threshold, exceed a predetermined web site traffic value. ### Claim 13 (depends on 12) The computer system of claim 12, wherein the machine learning algorithm utilizes supervised learning, unsupervised learning, reinforcement learning and/or semi-supervised learning; and wherein the neural network is multilayered and includes an input layer and one or more of a hidden layer and an output layer. ### Claim 14 (depends on 8) The computer system of claim 8, wherein the one or more action recommendations received from the digital content management system is fed into the neural network to further train the neural network. ### Claim 15 A non-transitory computer-readable medium for executing a machine learning algorithm for generating optimized digital content, comprising: a memory having processor-readable instructions stored therein, to direct a processor for: receiving, at a prediction engine of a server, one or more digital content items from one or more content partners, the one or more digital content items including performance information; analyzing, using a machine learning algorithm and a neural network, each of the one or more digital content items and against performance criteria to determine a corresponding potential total traffic value; comparing, by the prediction engine, the corresponding potential total traffic value for each of the one or more digital content items against a traffic threshold; based on comparing the corresponding potential total traffic value for each of the one or more digital content items against the traffic threshold, determining, by the prediction engine, a first subset of the one or more digital content items and a second subset of the one or more digital content items, wherein the first subset of the one or more digital content items include the one or more digital content items that have met or exceeded the traffic threshold, and the second subset of the one or more digital content items include the one or more digital content items that have not met the traffic threshold; transmitting, by a decision engine of the server, the first subset of the one or more digital content items that have met or exceeded the traffic threshold to a digital content management system; generating, by the decision engine of the server, a ranking of the first subset of the one or more digital content items and a corresponding action recommendation determined by the machine learning algorithm and the neural network; and conveying to a user, by the decision engine of the server, the ranking of the first subset of the one or more digital content items and the corresponding action recommendation as action buttons, the action buttons including a first button that instantly publishes a top emerging digital content item to mobile devices and a second button that provides insight information comprising expected performance. ### Claim 16 (depends on 15) The non-transitory computer-readable medium of claim 15, wherein training data of the neural network is trained over a predetermined period of time. ### Claim 17 (depends on 15) The non-transitory computer-readable medium of claim 15, wherein generating a ranking and a corresponding action recommendation comprises using the machine learning algorithm to compare the one or more digital content items to editorial criteria, the editorial criteria comprising at least one of metrics and features. ### Claim 18 (depends on 15) The non-transitory computer-readable medium of claim 15, wherein the machine learning algorithm utilizes the neural network to forecast digital content item performance. ### Claim 19 (depends on 18) The non-transitory computer-readable medium of claim 18, wherein the machine learning algorithm utilizes supervised learning, unsupervised learning, reinforcement learning and/or semi-supervised learning; and wherein the neural network is multilayered and includes an input layer and one or more of a hidden layer and an output layer. ### Claim 20 (depends on 15) The non-transitory computer-readable medium of claim 15, wherein the one or more action recommendations received from the digital content management system is fed into the neural network to further train the neural network.