# US 10,311,372 B1 · Machine learning based content delivery > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,311,372 B1 - Document ID: US10311372B1 - Status: Granted - Kind: B1 - Application number: 14/578,099 - Filed: 2014-12-19 - Granted: 2019-06-04 - Earliest claimed: 2014-12-19 - Assignees: Amazon Technologies, Inc. - Applicants: Amazon Technologies, Inc. - Inventors: Blair Livingstone Hotchkies; Bradley Scott Bowman; Paul Christopher Cerda; Min Chong; Anthony T. Chor; Leo Parker Dirac; Kevin Andrew Granade; Udip Pant; Sean Michael Scott - Web page: https://patentgrep.com/patent/US10311372B1 ## CPC - G06N 20/00: Machine learning - G06N 3/04: Architecture, e.g. interconnection topology - G06N 3/0499: Feedforward networks - G06N 3/09: Supervised learning - G06N 5/04: Inference or reasoning models - H04L 67/02: based on web technology, e.g. hypertext transfer protocol [HTTP] - 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] - H04L 67/2885: Hierarchically arranged intermediate devices, e.g. for hierarchical caching - H04L 67/5681: Pre-fetching or pre-delivering data based on network characteristics - H04L 67/60: Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources (admission control or resource allocation H04L47/70) - 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 for managing content delivery functionalities based on machine learning models are provided. In one aspect, content requests are routed in accordance with clusters of historical content requests to optimize cache performance. In another aspect, content delivery strategies for responding to content requests are determined based on a model trained on data related to historical content requests. The model may also be used to determine above-the-fold configurations for rendering responses to content requests. In some embodiments, portions of the model can be executed on client computing devices. ## Claims ### Claim 1 A computer-implemented method for facilitating content delivery, the computer-implemented method comprising: under control of a hardware computing device configured with specific computer executable instructions, training a machine learning model that predicts content delivery performance based, at least in part, on historical data related to content requests over a specified period of time; providing a first separable portion of the machine learning model to a user computing device; obtaining an incoming content request from the user computing device, wherein the incoming content request includes information generated by the first separable portion of the machine learning model executing on the user computing device; identifying a predicted content delivery performance that results from an application of a second separable portion of the machine learning model to a candidate content delivery strategy for configuring a content page corresponding to the incoming content request in a first above-the-fold (ATF) configuration; and in response to a determination that the predicted content delivery performance satisfies a predetermined condition: configuring the requested content page in the first ATF configuration; generating a response to the incoming content request in accordance with configuring the requested content page in the first ATF configuration, wherein the generated response indicates a priority for retrieving network resources to be displayed in an ATF portion of the requested content page; and transmitting the generated response to the user computing device, wherein reception of the generated response causes the user computing device to retrieve the network resources in accordance with the priority indicated in the generated response. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein the first separable portion of the machine learning model is trained on the user computing device. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein identifying the predicted content delivery performance comprises providing the information generated by the first separable portion of the machine learning model into the second separable portion of the machine learning model. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein the first separable portion of the machine learning model receives input data from the user computing device. ### Claim 5 (depends on 4) The computer-implemented method of claim 4, wherein at least one portion of the input data is derived from information accessible exclusively to the user computing device. ### Claim 6 A computer-implemented method comprising: under control of a hardware computing device configured with specific computer executable instructions, obtaining a machine learning model trained to facilitate content delivery in response to content requests; causing a first separable portion of the machine learning model to be provided to a user computing device; in response to a first content request submitted by the user computing device, identifying a first content delivery strategy for configuring a content page corresponding to the first content request in a first above-the-fold (ATF) configuration as a result of the first separable portion of the machine learning model executing on the user computing device; and configuring the content page in the first ATF configuration and indicating a priority for retrieving network resources to be displayed in an ATF portion of the content page. ### Claim 7 (depends on 6) The computer-implemented method of claim 6 further comprising updating the first separable portion of the machine learning model based, at least in part, on content delivery performance data related to the implementation of the first delivery strategy. ### Claim 8 (depends on 7) The computer-implemented method of claim 7 further comprising causing the updated first separable portion of the machine learning model to be provided to the user computing device. ### Claim 9 (depends on 8) The computer-implemented method of claim 8 further comprising, in response to a second content request submitted by the user computing device, identifying a second content delivery strategy as a result of the updated first separable portion of the machine learning model executing on the user computing device. ### Claim 10 (depends on 6) The computer-implemented method of claim 6, wherein the machine learning model corresponds to a decision tree model or an artificial neural network model. ### Claim 11 A system comprising: a data store configured to at least store computer-executable instructions; and a hardware processor in communication with the data store, the hardware processor configured to execute the computer-executable instructions to at least: obtain a machine learning model trained to facilitate content delivery in response to content requests; cause a first separable portion of the machine learning model to be provided to a user computing device; in response to a first content request submitted by the user computing device, identify a first content delivery strategy for configuring a content page corresponding to the first content request in a first above-the-fold (ATF) configuration as a result of the first separable portion of the machine learning model executing on the user computing device; and configure the content page in the first ATF configuration and indicating a priority for retrieving network resources to be displayed in an ATF portion of the content page. ### Claim 12 (depends on 11) The system of claim 11, wherein the hardware processor is further configured to cause a second separable portion of the machine learning model to be provided to the user computing device. ### Claim 13 (depends on 12) The system of claim 12, wherein the second separable portion of the machine learning model is selected based, at least in part, on information related to the first content request. ### Claim 14 (depends on 12) The system of claim 12, wherein the hardware processor is further configured to, in response to a second content request submitted by the user computing device, identify a second content delivery strategy as a result of the second separable portion of the machine learning model executing on the user computing device. ### Claim 15 (depends on 12) The system of claim 12, wherein the implementation of the first content delivery strategy includes causing the second separable portion of the machine learning model to be provided to the user computing device machine. ### Claim 16 A non-transitory computer readable storage medium storing computer executable instructions that when executed by a processor perform operations comprising: obtaining a machine learning model trained to facilitate content delivery in response to content requests; causing a first separable portion of the machine learning model to be provided to a first user computing device; in response to a first content request submitted by the first user computing device, identifying a first content delivery strategy for configuring a content page corresponding to the first content request in a first above-the-fold (ATF) configuration as a result of the first separable portion of the machine learning model executing on the first user computing device; and configuring the content page in the first ATF configuration and indicating a priority for retrieving network resources to be displayed in an ATF portion of the content page. ### Claim 17 (depends on 16) The non-transitory computer-readable storage medium of claim 16, wherein the operations further comprise causing a second separable portion of the machine learning model to be provided to a second user computing device. ### Claim 18 (depends on 17) The non-transitory computer-readable storage medium of claim 17, wherein the first separable portion of the machine learning model includes the second separable portion. ### Claim 19 (depends on 16) The non-transitory computer-readable storage medium of claim 16, wherein the operations further comprise obtaining updates to the machine learning model. ### Claim 20 (depends on 16) The non-transitory computer-readable storage medium of claim 16, wherein the operations further comprises causing updates to the first separable portion of the machine learning model to be provided to the first user computing device.