# US 9,436,754 B1 · Unit group generation and relationship establishment > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 9,436,754 B1 - Document ID: US9436754B1 - Status: Granted - Kind: B1 - Application number: 13/934,813 - Filed: 2013-07-03 - Granted: 2016-09-06 - Earliest claimed: 2013-07-03 - Assignees: Pinterest, Inc. - Applicants: Pinterest, Inc. - Inventors: Ningning Hu; Tze Way Eugene Ie - Web page: https://patentgrep.com/patent/US9436754B1 ## CPC - G06F 16/285: {Clustering or classification} - G06F 16/907: Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract This disclosure describes systems and methods for establishing a unit group dictionary based on user provided annotations. The unit group dictionary may be used to identify relationships between multiple items in a corpus. Those relationships may facilitate the display of object identifiers and/or other aspects used and/or provided by the object management service. ## Claims ### Claim 1 A computer-implemented method, comprising: under control of one or more computing systems configured with executable instructions, identifying a corpus of items, wherein a plurality of the corpus of items include annotations; processing the annotations to identify a plurality of non-random unit groups; for each of the plurality of non-random unit groups, determining a discount factor based at least in part on a quotient of a number of times the non-random unit group appears in the corpus and a number of items in the corpus; and generating a unit group dictionary that includes non-random unit groups with a discount factor above a defined threshold. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein each of the plurality of non-random unit groups include at least two units. ### Claim 3 (depends on 2) The computer-implemented method of claim 2, wherein each of the plurality of non-random unit groups are identified based at least in part on a probability that the at least two units of the unit group appearing together in the corpus is greater than a probability of the at least two units of the unit group appearing independently in the corpus. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein the items include at least one of a representation, an object identifier, a set, or a user. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, wherein a non-random unit group is identified in the annotations based at least in part on a probability that units of the unit group appear together in the corpus is greater than a probability of the units appearing separately in the corpus. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, wherein the units include at least one of a character, a word, a symbol, or a number. ### Claim 7 (depends on 1) The computer-implemented method of claim 1, wherein the corpus includes all items managed by an object management service. ### Claim 8 (depends on 1) The computer-implemented method of claim 1, wherein the discount factor is a log of the quotient. ### Claim 9 A non-transitory computer-readable storage medium storing instructions to establish relationships between items, the instructions when executed by a processor causing the processor to at least: identify a unit group for a first item, wherein the unit group is established based at least in part on a user provided annotation; identify a second item of a corpus of items having the unit group associated therewith; establish a relationship between the first item and the second item based at least in part on the unit group; determine a discount factor for the unit group based at least in part on a quotient of a frequency of the unit group in a corpus and a number of items in the corpus; receive a request from a computing device to view the first item; and send for display the first item and the second item. ### Claim 10 (depends on 9) The non-transitory computer-readable storage medium of claim 9, wherein the discount factor is a log of the quotient. ### Claim 11 (depends on 9) The non-transitory computer-readable storage medium of claim 9, wherein the unit group includes at least one of a character, a word, a symbol or a number. ### Claim 12 (depends on 9) The non-transitory computer-readable storage medium of claim 9, wherein the unit group is selected from a plurality of unit groups defined for the corpus. ### Claim 13 (depends on 9) The non-transitory computer-readable storage medium of claim 9, wherein the unit group for the first item is identified based on a selection of a unit group from a plurality of unit groups associated with the first item. ### Claim 14 (depends on 9) The non-transitory computer-readable storage medium of claim 9, wherein the relationship is further established based at least in part on the first item and the second item having a same unit group. ### Claim 15 A computing system, comprising: one or more processors; and a memory coupled to the one or more processors and storing program instructions that when executed by the one or more processors cause the one or more processors to: identify a corpus including a plurality of items, wherein each of the plurality of items include associated annotations; process the associated annotations to identify a first plurality of non-random unit groups, wherein each non-random unit group includes at least two units included in the associated annotations; for each non-random unit group of the first plurality of non-random unit groups, generate a discount factor based at least in part on a quotient of a frequency of the unit group in the corpus and a number of the plurality of items in the corpus; and generate a dictionary of a second plurality of non-random unit groups identified from the first plurality of non-random unit groups. ### Claim 16 (depends on 15) The computing system of claim 15, wherein each of the first plurality of non-random unit groups are determined based on a joint probability that the at least two units of the unit group appearing together in the corpus is greater than a probability of the at least two units of the unit group appearing independently in the corpus. ### Claim 17 (depends on 16) The computing system of claim 16, wherein the joint probability is determined based at least in part on a number of times the at least two units of the unit group appear together in the corpus. ### Claim 18 (depends on 16) The computing system of claim 16, wherein the probability is determined based at least in part on a number of times the at least two units of the unit group appear independently in the corpus. ### Claim 19 (depends on 15) The computing system of claim 15, wherein the discount factor is a log of the quotient. ### Claim 20 (depends on 15) The computing system of claim 15, wherein the corpus includes all items managed by an object management service.