# US 2020/0401666 A1 · Systems and methods for classifying content items based on user comments > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2020/0401666 A1 - Document ID: US20200401666A1 - Status: Published application - Kind: A1 - Application number: 15/955,144 - Filed: 2018-04-17 - Published: 2020-12-24 - Earliest claimed: 2018-04-17 - Assignees: Facebook, Inc. - Applicants: Facebook, Inc. - Inventors: Umut Ozertem; Ying Zhang - Web page: https://patentgrep.com/patent/US20200401666A1 ## CPC - G06F 16/287: {Visualization; Browsing} - G06F 16/955: using information identifiers, e.g. uniform resource locators [URL] - G06N 20/00: Machine learning - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Systems, methods, and non-transitory computer readable media can determine a relationship type between a first content item and a second content item based on a comment associated with at least one of the first content item and the second content item. A machine learning model can be trained based on the first content item, the second content item, and the determined relationship type. A related content item of a content item can be determined based on the machine learning model. ## Claims ### Claim 1 A computer-implemented method comprising: determining, by a computing system, a relationship type between a first content item and a second content item based on a comment associated with at least one of: the first content item or the second content item, wherein the determined relationship type is at least one of a plurality of relationship types including a similar viewpoint, a different viewpoint, and an alternative viewpoint; training, by the computing system, a machine learning model based on the first content item, the second content item, and the determined relationship type; and determining, by the computing system, a related content item of a content item by the machine learning model based on a ranking of the plurality of relationship types between the related content item and the content item. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein the first content item and the second content item are determined to be related based on a social graph associated with a social networking system. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein the first content item and the second item include one or more of: an article, a link, or a uniform resource locator (URL). ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein the first content item is included in a post and the second content item is included in a comment associated with the post. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, wherein a link to the first content item is included in a post and a link to the second content item is included in a comment associated with the post. ### Claim 6 (depends on 5) The computer-implemented method of claim 5, wherein the relationship type between the first content item and the second content item is determined based on analyzing content of the comment associated with the post. ### Claim 7 (depends on 1) The computer-implemented method of claim 1, further comprising providing the related content item in a feed of a user. ### Claim 8 (depends on 1) The computer-implemented method of claim 1, wherein the content item and the related content item are articles, and wherein the method further comprises: extracting statements in the content item and the related content item; and generating a summary based on overlapping statements of the extracted statements. ### Claim 9 (depends on 1) The computer-implemented method of claim 1, wherein the determining a related content item of the content item based on the machine learning model includes: receiving the content item and an input relationship type as input to the machine learning model; and providing the related content item as output of the machine learning model based on the ranking of the plurality of relationship types between the related content item and the content item, wherein the ranking indicates a likelihood of the content item and the related content item having the input relationship type. ### Claim 10 (depends on 1) The computer-implemented method of claim 1, further comprising: receiving related content items as input to the machine learning model; and providing an output relationship type between the related content items as output of the machine learning model. ### Claim 11 A system comprising: at least one hardware processor; and a memory storing instructions that, when executed by the at least one processor, cause the system to perform: determining a relationship type between a first content item and a second content item based on a comment associated with at least one of the first content item and the second content item, wherein the determined relationship type is at least one of a plurality of relationship types including a similar viewpoint, a different viewpoint, and an alternative viewpoint; training a machine learning model based on the first content item, the second content item, and the determined relationship type; and determining a related content item of a content item by the machine learning model based on a ranking of the plurality of relationship types between the related content item and the content item. ### Claim 12 (depends on 11) The system of claim 11, wherein the first content item and the second content item are determined to be related based on a social graph associated with a social networking system. ### Claim 13 (depends on 11) The system of claim 11, wherein a link to the first content item is included in a post and a link to the second content item is included in a comment associated with the post. ### Claim 14 (depends on 13) The system of claim 13, wherein the relationship type between the first content item and the second content item is determined based on analyzing content of the comment associated with the post. ### Claim 15 (depends on 11) The system of claim 11, wherein the content item and the related content item are articles, and the instructions further cause the system to perform: extracting statements in the content item and the related content item; and generating a summary based on overlapping statements of the extracted statements. ### Claim 16 A non-transitory computer readable medium including instructions that, when executed by at least one hardware processor of a computing system, cause the computing system to perform a method comprising: determining a relationship type between a first content item and a second content item based on a comment associated with at least one of the first content item or the second content item, wherein the determined relationship type is at least one of a plurality of relationship types including a similar viewpoint, a different viewpoint, and an alternative viewpoint; training a machine learning model based on the first content item, the second content item, and the determined relationship type; and determining a related content item of a content item by the machine learning model based on a ranking of the plurality of relationship types between the related content item and the content item. ### Claim 17 (depends on 16) The non-transitory computer readable medium of claim 16, wherein the first content item and the second content item are determined to be related based on a social graph associated with a social networking system. ### Claim 18 (depends on 16) The non-transitory computer readable medium of claim 16, wherein a link to the first content item is included in a post and a link to the second content item is included in a comment associated with the post. ### Claim 19 (depends on 18) The non-transitory computer readable medium of claim 18, wherein the relationship type between the first content item and the second content item is determined based on analyzing content of the comment associated with the post. ### Claim 20 (depends on 16) The non-transitory computer readable medium of claim 16, wherein the content item and the related content item are articles, and wherein the method further comprises: extracting statements in the content item and the related content item; and generating a summary based on overlapping statements of the extracted statements.