User interface for identifying top attributes
In various example embodiments, a system and method for identifying top attributes are presented. Text content from an electronic document describing a product is extracted. A plurality of attributes associated with the product are identified using the extracted text content. A descriptor word proximate to each of the plurality of attributes in the extracted text content is detected. A weight is assigned to each descriptor word proximate to each of the plurality of attributes. The plurality of attributes are ranked based on the weight of the descriptor word proximate to each of the plurality of attributes. A predetermined number of top ranked attributes from the plurality of attributes are caused to be presented. Further, an image associated with each of the top ranked attributes is determined, the image being designated as a visual representation of the corresponding top ranked attribute. The image is caused to be displayed in association with the product on a user interface.
1. A system comprising:
one or more processors and a memory including instructions, which when executed by the one or more processors, cause the one or more processors to perform operations comprising:
extracting text content from an electronic document describing a product;
using the extracted text content, identifying a plurality of attributes associated with the product;
detecting at least one descriptor word proximate to each of the plurality of attributes in the extracted text content;
assigning corresponding weights to each descriptor word proximate to each of the plurality of attributes;
ranking the plurality of attributes based on the weight of the at least one descriptor word proximate to each of the plurality of attributes; and
causing generation of a user interface that includes a number of images, each image within the number of images being a visual representation of a top ranked attribute selected based on the ranking of the plurality of attributes.
2. The system of claim 1 , wherein the identifying the plurality of attributes comprises:
determining, from the database, a category of the product; and
identifying attributes in the extracted text content that match a set of attributes stored in the database for the category of the product.
3. The system of claim 1 , wherein the extracted text content is associated with a product description or user comment on the product.
4. The system of claim 1 , wherein the at least one descriptor word occurs immediately before an attribute of the plurality of attributes.
5. The system of claim 1 , wherein the at least one descriptor word is an adjective describing an attribute of the plurality of attributes.
6. The system of claim 1 , wherein the ranking comprises:
comparing a statistical measurement of the weight assigned to each descriptor word to a threshold measurement to determine an outlier descriptor word; and
removing the outlier descriptor word.
7. The system of claim 1 , wherein assigning weights to each descriptor word proximate to each of the plurality of attributes comprises assigning a weight to each instance of a descriptor word proximate to each attribute, wherein, when there are multiple instances of a particular descriptor word being proximate to a particular attribute within the text content, weights for each instance of the particular descriptor word are used in ranking the particular attribute.
8. The system of claim 1 , wherein causing generation of the user interface comprises causing each image within the number of images to be superimposed over a description section associated with the product on the user interface.
9. The system of claim 1 , wherein causing generation of the user interface comprises causing each image within the number of images to be superimposed over a product image of the product on the user interface.
10. The system of claim 1 , wherein each image within the number of images is associated with a selectable element within the user interface and wherein the operations further comprise:
in response to a selection of at least one selectable element, determining other products having the attribute represented by the image associated with the at least one selectable element as a top ranked attribute; and
causing presentation of a user interface that includes information on the other products.
11. A computerized method performed by one or more processors and comprising:
extracting text content from an electronic document describing a product;
using the extracted text content, identifying a plurality of attributes associated with the product;
detecting at least one descriptor word proximate to each of the plurality of attributes in the extracted text content;
assigning corresponding weights to each descriptor word proximate to each of the plurality of attributes;
ranking the plurality of attributes based on the weight of the at least one descriptor word proximate to each of the plurality of attributes; and
causing generation of a user interface that includes a number of images, each image within the number of images being a visual representation of a top ranked attribute selected based on the ranking of the plurality of attributes.
12. The method of claim 11 , wherein the identifying the plurality of attributes comprises:
determining, from the database, a category of the product; and
identifying attributes in the extracted text content that match a set of attributes stored in the database for the category of the product.
13. The method of claim 11 , wherein the extracted text content is associated with a product description or user comment on the product.
14. The method of claim 11 , wherein the at least one descriptor word occurs immediately before an attribute of the plurality of attributes.
15. The method of claim 11 , wherein the at least one descriptor word is an adjective describing an attribute of the plurality of attributes.
16. The method of claim 11 , wherein the ranking comprises:
comparing a statistical measurement of the weight assigned to each descriptor word to a threshold measurement to determine an outlier descriptor word; and
removing the outlier descriptor word.
17. The method of claim 11 , wherein assigning weights to each descriptor word proximate to each of the plurality of attributes comprises assigning a weight to each instance of a descriptor word proximate to each attribute, wherein, when there are multiple instances of a particular descriptor word being proximate to a particular attribute within the text content, weights for each instance of the particular descriptor word are used in ranking the particular attribute.
18. The method of claim 11 , wherein causing generation of the user interface comprises causing each image within the number of images to be superimposed over a description section associated with the product on the user interface.
19. The method of claim 11 , wherein causing generation of the user interface comprises superimposing the causing each image within the number of images to be superimposed over a product image of the product on the user interface.
20. A machine-readable medium having no transitory signals and storing instructions that, when executed by at least one processor of a machine, cause the machine to perform operations comprising:
extracting text content from an electronic document describing a product;
using the extracted text content, identifying a plurality of attributes associated with the product;
detecting at least one descriptor word proximate to each of the plurality of attributes in the extracted text content;
assigning corresponding weights to each descriptor word proximate to each of the plurality of attributes;
ranking the plurality of attributes based on the weight of the at least one descriptor word proximate to each of the plurality of attributes; and
causing generation of a user interface that includes a number of images, each image within the number of images being a visual representation of a top ranked attribute selected based on the ranking of the plurality of attributes.