Abstract: Described herein are embodiments for machine-generating and naming ontologies for for-sale items. A neural network may be used to train information describing for-sale items into feature vectors that describe the for-sale items. These feature vectors may be sorted into clusters based on their relative proximity using clustering algorithms. The resulting clusters may be sub-divided into smaller clusters depending on the precision used in the clustering algorithm. The set of clusters may form a hierarchical data structure where each level has clusters determined at a certain precision and each lower level sub-divides those clusters. The clusters may be named based on the most salient facets that describe the for-sale items in the clusters.

Abstract: Disclosed herein are system, computer-readable storage medium, and method embodiments of automatic ontology generation by embedding representations. A system including at least one processor may be configured to receive a vectorized feature set derived from an embedding and including first and second features, and provide the vectorized feature set to a fuser set including first and second fusers. The system may be configured to generate a representation from the fuser set based on the first and second features, and derive tasks based on the representation, assigning to the tasks respective qualifier sets including a weight value, a loss function, and a feedforward function. The system may be configured to compute respective weighted losses for the tasks, based on the respective qualifier sets, and output a data model based on backpropagating the respective weighted losses through the fuser set, the vectorized feature set, the embedding, or a combination thereof.

Abstract: Described herein are embodiments for improving search engine results of listings of For Sale Objects (FSOs). A search engine may be improved by implementing rules that resolve ambiguity between listings for different (FSOs) that match the same search inputs. An unsupervised machine learning module may evaluate candidate rules and identify improvements that may not be obvious to a human evaluator. An ecommerce site that combines the improved search engine with the unsupervised machine learning module may dynamically evaluate search results using different candidate rules and iteratively improve search results.

Abstract: Described herein are embodiments for assisting in creating a listing for a For Sale Object (FSO). An item name suggestion module receives seller input and provides suggested entries for the listing to help the seller describe the FSO more accurately and consistently. A hierarchical database provides a structure for ordering suggested entries, with the structure ordered based on scores. The scores are based on rules that relate item characteristics and take into account rankings of those item characteristics with respect to one another. Metadata tags that are used by the online merchandise platforms can be identified and included in the listing, even if a seller is not familiar with the metadata tags. The hierarchical database also connects or associates item characteristics in groups that describe specific FSO. The connections can help to optimize search results as the listing is completed by the seller.