Taste extraction curation and tagging
In non-limiting examples of the present disclosure, taste data is generated and usable within one or more applications. A taste is one or more elements that describe an entity. Information of an application may be processed to extract entity data that corresponds to a plurality of candidates to be designated as tastes (taste data). The candidates for the tastes are curated. In examples, the curating comprises filtering the candidates for tastes to remove extracted candidates. Extraction rules for managing structured taste data may be applied in the filtering. A status of a remaining candidate may be determined as approved or rejected based on processing of received user feedback. Taste data may be generated for an approved candidate. The generating of the taste data may comprise assigning parameters that include a descriptor type and a recommendation type. Taste data may be used within an application.
1. A computer-implemented method comprising:
processing free-form data received by an application to identify entity data;
upon identifying the entity data, extracting one or more instances of free-from data associated with the entity data that corresponds to a plurality of candidates to be designated as tastes, wherein a taste is one or more elements that describe an entity;
curating the candidates, wherein curating comprises:
filtering the plurality candidates to remove a subset of candidates based on application of extraction rules for managing taste data stored in a memory;
determining at least one approved candidate, wherein the determination is determined based at least upon received user feedback; and
generating taste data for the approved candidate, wherein the generating the taste data comprises assigning parameters that comprise a descriptor type and a recommendation type;
associating the generated taste data with content of the application based at least upon analyzing the assigned parameters of the taste data and attributes of the content;
generating structured taste data based upon the association of the generated taste data with the content, wherein the structured taste data stores one or more tastes and associations between the one or more tastes and the content such that the one or more tastes can be presented with the content; and
causing the application to present the one or more tastes with associated application content.
2. The computer-implemented method according to claim 1 , wherein causing the application to present the one or more tastes further comprises causing the display of the one or more tastes within the application as directed information.
3. The computer-implemented method according to claim 1 , wherein curating further comprises clustering the generated taste data in a cluster with other taste data, the cluster representing hierarchical relationships between the generated taste data and other taste data, and storing cluster data for the cluster in the memory to update the structured taste data.
4. The computer-implemented method according to claim 3 , wherein generating the taste data further comprises identifying whether the generated taste data is a non-compositional compound, and the clustering further comprises managing implications associated with clustering the generated taste data and the other taste data in response to identifying whether the generated taste data is the non-compositional compound.
5. The computer-implemented method according to claim 3 , wherein clustering further comprises associating synonyms for the generated taste data in the cluster based on matching the generated taste data with the other taste data.
6. The computer-implemented method according to claim 5 , wherein clustering further comprises determining a canonical phrase to represent the cluster data and setting any of the cluster data to be presentable in the application as the canonical phrase.
7. The computer-implemented method according to claim 6 , wherein clustering further comprises synthesizing the cluster data, wherein the synthesizing comprises at least one of rephrasing a portion of the cluster data, and expanding the cluster to include created taste data.
8. The computer-implemented method according to claim 7 , wherein associating of the generated taste data with the content further comprises associating the content with the canonical phrase, and wherein the computer-implemented method further comprising presenting the canonical phrase to a user of the application based on the stored associations, and wherein the presenting presents the canonical phrase within the application in association with creation of directed information.
9. The computer-implemented method according to claim 1 , wherein associating further comprises applying data association rules to determine associations between the generated taste data and the content, and wherein the data association rules comprise determining whether to blacklist generated taste data from being associated with particular content and selectively blacklisting the generated taste data.
10. A system comprising:
at least one processor; and
a memory operatively connected with the at least one processor, the memory comprising computer executable instructions that, when executed by the at least one processor, perform a method comprising:
processing free-form data received by an application to identify entity data;
upon identifying the entity data, extracting one or more instances of free-from data associated with the entity data that corresponds to a plurality of candidates to be designated as tastes, wherein a taste is one or more elements that describe an entity;
curating the candidates, wherein curating comprises:
filtering the plurality candidates to remove a subset of candidates based on application of extraction rules for managing taste data stored in a memory;
determining at least one approved candidate, wherein the determination is determined based at least upon received user feedback; and
generating taste data for the approved candidate, wherein the generating the taste data comprises assigning parameters that comprise a descriptor type and a recommendation type;
associating the generated taste data with content of the application based on analyzing the assigned parameters of the taste data and attributes of the content;
generating structured taste data based upon the association of the generated taste data with the content, wherein the structured taste data stores one or more tastes and associations between the one or more tastes and the content such that the one or more tastes can be presented with the content; and
causing the application to present the one or more tastes with associated application content.
11. The system according to claim 10 , wherein causing the application to present the one or more tastes further comprises causing the comprising presenting the taste display of the one or more tastes within the application as directed information.
12. The system according to claim 11 , wherein curating further comprises clustering the generated taste data in a cluster with other taste data, the cluster representing hierarchical relationships between the generated taste data and other taste data, and storing cluster data for the cluster in the storage to update the structured taste data.
13. The system according to claim 12 , wherein generating of the taste data further comprises identifying whether the generated taste data is a non-compositional compound and the clustering further comprises managing implications associated with the cluster data in response to identifying whether the generated taste data is the non-compositional compound.
14. The system according to claim 13 , wherein clustering further comprises associating synonyms for the generated taste data in the cluster based on matching the generated taste data with the other taste data.
15. The system according to claim 14 , wherein clustering further comprises determining a canonical phrase to represent the cluster data and setting any of the cluster data to be presentable in the application as the canonical phrase.
16. The system according to claim 15 , wherein clustering further comprises synthesizing the cluster data, wherein the synthesizing comprises at least one of rephrasing a portion of the cluster data, and expanding the cluster to include created taste data.
17. The system according to claim 16 , wherein associating of the generated taste data with the content further comprises associating the content with the canonical phrase, and wherein the method further comprising presenting the canonical phrase to a user of the application based on the stored associations, and wherein the presenting displays the canonical phrase within the application in association with creation of directed information.
18. The system according to claim 10 , wherein associating further comprises applying data association rules to determine associations between the generated taste data and the content, and wherein the data association rules comprise determining whether to blacklist generated taste data from being associated with particular content and selectively blacklisting the generated taste data.
19. A computer-readable medium including executable instructions, that when executed by at least one processor, causing the processor to perform operations comprising:
processing free-form data received by an application to identify entity data;
upon identifying the entity data, extracting one or more instances of free-from data associated with the entity data that corresponds to a plurality of candidates to be designated as tastes, wherein a taste is one or more elements that describe an entity;
curating the candidates, wherein curating comprises:
filtering the plurality candidates to remove a subset of candidates based on application of extraction rules for managing taste data stored in a memory;
determining at least one approved candidate, wherein the determination is determined based at least upon received user feedback; and
generating taste data for the approved candidate, wherein the generating the taste data comprises assigning parameters that comprise a descriptor type and a recommendation type;
associating the generated taste data with content of the application based at least upon analyzing the assigned parameters of the taste data and attributes of the content;
generating structured taste data based upon the association of the generated taste data with the content, wherein the structured taste data stores one or more tastes and associations between the one or more tastes and the content such that the one or more tastes can be presented with the content; and
causing the application to present the one or more tastes with associated application content.
20. The computer storage medium of claim 19 , wherein associating further comprises applying data association rules to determine associations between the generated taste data and the content, and wherein the data association rules comprise determining whether to blacklist generated taste data from being associated with particular content and selectively blacklisting the generated taste data.