Product recommendation with ontology-linked product review
Techniques for product recommendation involve receiving, from a user, a request for recommendation of a product matching one or more characteristics specified in the request. In response to the request, a product may be identified for recommendation, at least in part by searching an ontology in which the identified product is represented as matching at least one of the characteristics specified in the request. The ontology may further include at least one link to at least one natural language product review providing supporting evidence that the identified product matches the at least one characteristic. At least a portion of the at least one natural language product review may be retrieved using the at least one link in the ontology, and presented to the user in support of a recommendation of the identified product, in response to the request.
1. A method comprising:
locating, by a document analyzer of a question answering (QA) system, a plurality of natural language product reviews about a first product by scraping a plurality of web pages;
augmenting, by a domain modeler of the QA system, an ontology, by including in the ontology links to the plurality of natural language product reviews located by scraping the plurality of web pages, wherein the ontology further includes a first product concept node representing the first product and one or more related concept nodes each representing a characteristic of the first product, wherein the ontology further includes, for each of the one or more related concept nodes, a relationship between the first product concept node and the related concept node, the relationship being encoded as a data item in the ontology;
receiving, by the QA system from a user, a request for recommendation of a product matching a characteristic specified in the request;
identifying, by the QA system, the first product for recommendation in response to the request, at least in part by searching the ontology for the characteristic specified in the request, wherein searching the ontology comprises matching the specified characteristic to a first related concept node of the one or more related concept nodes, the first related concept node representing the specified characteristic, and traversing the relationship in the ontology from the first related concept node of the specified characteristic to the first product concept node of the first product, wherein at least one of the plurality of natural language product reviews provides an indication that the first product matches the specified characteristic;
retrieving, by the QA system, at least portions of at least two of the plurality of natural language product reviews using the links in the ontology; and
presenting to the user, by the QA system within a user interface in response to the request, the retrieved at least portions of the at least two natural language product reviews in support of a recommendation of the identified first product.
2. The method of claim 1 , wherein the at least one of the plurality of natural language product reviews comprises an evaluation of the first product made by a purchaser of the first product.
3. The method of claim 1 , wherein the at least one of the plurality of natural language product reviews comprises an evaluation of the first product made based on use of the first product by an author of the at least one natural language product review.
4. The method of claim 1 , wherein the at least one of the plurality of natural language product reviews is linked in the ontology to the data item encoding the relationship between the first product concept node and the first related concept node.
5. The method of claim 1 , wherein retrieving the at least portions of the at least two natural language product reviews comprises following at least one link from the data item encoding the relationship in the ontology traversed to identify the first product for recommendation.
6. The method of claim 1 , wherein the characteristic specified in the request is a first characteristic, wherein identifying the first product for recommendation comprises determining, based at least in part on analysis of a first natural language product review, that the first product also matches a second characteristic specified in the request.
7. The method of claim 6 , further comprising updating the ontology to include a relationship between the first product and the second characteristic determined to match the first product based on the first natural language product review, and to include a link to the first natural language product review in association with the relationship between the first product and the second characteristic.
8. At least one non-transitory computer-readable storage medium storing computer-executable instructions that, when executed, perform a method comprising:
locating, by a document analyzer of a question answering (QA) system, a plurality of natural language product reviews about a first product by scraping a plurality of web pages;
augmenting, by a domain modeler of the QA system, an ontology, by including in the ontology links to the plurality of natural language product reviews located by scraping the plurality of web pages, wherein the ontology further includes a first product concept node representing the first product and one or more related concept nodes each representing a characteristic of the first product, wherein the ontology further includes, for each of the one or more related concept nodes, a relationship between the first product concept node and the related concept node, the relationship being encoded as a data item in the ontology;
receiving, by the QA system from a user, a request for recommendation of a product matching a characteristic specified in the request;
identifying, by the QA system, the first product for recommendation in response to the request, at least in part by searching the ontology for the characteristic specified in the request, wherein searching the ontology comprises matching the specified characteristic to a first related concept node of the one or more related concept nodes, the first related concept node representing the specified characteristic, and traversing the relationship in the ontology from the first related concept node of the specified characteristic to the first product concept node of the first product, wherein at least one of the plurality of natural language product reviews provides an indication that the first product matches the specified characteristic;
retrieving, by the QA system, at least portions of at least two of the plurality of natural language product reviews using the links in the ontology; and
presenting to the user, by the QA system within a user interface in response to the request, the retrieved at least portions of the at least two natural language product reviews in support of a recommendation of the identified first product.
9. The at least one non-transitory computer-readable storage medium of claim 8 , wherein the at least one of the plurality of natural language product reviews comprises an evaluation of the first product made by a purchaser of the first product.
10. The at least one non-transitory computer-readable storage medium of claim 8 , wherein the at least one of the plurality of natural language product reviews comprises an evaluation of the first product made based on use of the first product by an author of the at least one natural language product review.
11. The at least one non-transitory computer-readable storage medium of claim 8 , wherein the at least one of the plurality of natural language product reviews is linked in the ontology to the data item encoding the relationship between the first product concept node and the first related concept node.
12. The at least one non-transitory computer-readable storage medium of claim 8 , wherein retrieving the at least portions of the at least two natural language product reviews comprises following at least one link from the data item encoding the relationship in the ontology traversed to identify the first product for recommendation.
13. The at least one non-transitory computer-readable storage medium of claim 8 , wherein the characteristic specified in the request is a first characteristic, wherein identifying the first product for recommendation comprises determining, based at least in part on analysis of a first natural language product review, that the first product also matches a second characteristic specified in the request.
14. The at least one non-transitory computer-readable storage medium of claim 13 , wherein the method further comprises updating the ontology to include a relationship between the first product and the second characteristic determined to match the first product based on the first natural language product review, and to include a link to the first natural language product review in association with the relationship between the first product and the second characteristic.
15. Apparatus comprising:
at least one processor; and
at least one storage medium storing processor-executable instructions that, when executed by the at least one processor, perform a method comprising:
locating, by a document analyzer of a question answering (QA) system, a plurality of natural language product reviews about a first product by scraping a plurality of web pages;
augmenting, by a domain modeler of the QA system, an ontology, by including in the ontology links to the plurality of natural language product reviews located by scraping the plurality of web pages, wherein the ontology further includes a first product concept node representing the first product and one or more related concept nodes each representing a characteristic of the first product, wherein the ontology further includes, for each of the one or more related concept nodes, a relationship between the first product concept node and the related concept node, the relationship being encoded as a data item in the ontology;
receiving, by the QA system from a user, a request for recommendation of a product matching a characteristic specified in the request;
identifying, by the QA system, the first product for recommendation in response to the request, at least in part by searching the ontology for the characteristic specified in the request, wherein searching the ontology comprises matching the specified characteristic to a first related concept node of the one or more related concept nodes, the first related concept node representing the specified characteristic, and traversing the relationship in the ontology from the first related concept node of the specified characteristic to the first product concept node of the first product, wherein at least one of the plurality of natural language product reviews provides an indication that the first product matches the specified characteristic;
retrieving, by the QA system, at least portions of at least two of the plurality of natural language product reviews using the links in the ontology; and
presenting to the user, by the QA system within a user interface in response to the request, the retrieved at least portions of the at least two natural language product reviews in support of a recommendation of the identified first product.
16. The apparatus of claim 15 , wherein the at least one of the plurality of natural language product reviews comprises an evaluation of the first product made by a purchaser of the first product.
17. The apparatus of claim 15 , wherein the at least one of the plurality of natural language product reviews comprises an evaluation of the first product made based on use of the first product by an author of the at least one natural language product review.
18. The apparatus of claim 15 , wherein retrieving the at least portions of the at least two natural language product reviews comprises following at least one link from the data item encoding the relationship in the ontology traversed to identify the first product for recommendation.
19. The apparatus of claim 15 , wherein the characteristic specified in the request is a first characteristic, wherein identifying the first product for recommendation comprises determining, based at least in part on analysis of a first natural language product review, that the first product also matches a second characteristic specified in the request.
20. The apparatus of claim 19 , the method further comprises updating the ontology to include a relationship between the first product and the second characteristic determined to match the first product based on the first natural language product review, and to include a link to the first natural language product review in association with the relationship between the first product and the second characteristic.