Abstract: In various example embodiments, a system and method for retrieving database records are presented. A method includes receiving an access request for database records, identifying a primary database record corresponding to the access request, and selecting a set of secondary database records corresponding to a portion of the access request. The method determines a set of differentiating elements within records of the set of secondary database records. Each differentiating element distinguishes at least one record of the set of secondary database records from one or more of the primary database record and a portion of the records included in the set of secondary database records. The method generates a formatted search interface including a set of selectable interface icons representing one or more differentiating elements and causes presentation of the formatted search interface at a client device associated with the access request.

Abstract: Systems and methods generate browse node pages. A combination of a category and one or more aspects is identified by analyzing historical user behavior data associated with a network-based system. A determination is made regarding whether existing browse node pages for the network-based system includes a browse node page for the identified combination. Each browse node page for the network-based system is accessible by a respective uniform resource location (URL) and provides a grouping of items in a respective category with one or more respective aspects. Based on a determination that a browse node page does not exist for the identified combination of the category and the one or more aspects, a browse node page for the identified combination is generated. The generated browse node page corresponds with a URL and provides items matching the identified combination of the category and the one or more aspects.

Abstract: Systems and methods generate browse node pages. A combination of a category and one or more aspects is identified by analyzing historical user behavior data associated with a network-based system. A determination is made regarding whether existing browse node pages for the network-based system includes a browse node page for the identified combination. Each browse node page for the network-based system is accessible by a respective uniform resource location (URL) and provides a grouping of items in a respective category with one or more respective aspects. Based on a determination that a browse node page does not exist for the identified combination of the category and the one or more aspects, a browse node page for the identified combination is generated. The generated browse node page corresponds with a URL and provides items matching the identified combination of the category and the one or more aspects.

Abstract: To improve the user experience, browse nodes for popular combinations of categories and aspects are desirable. When a particular combination of category and aspects is found to have been requested at a frequency that exceeds a threshold, the existence of a browse node for that combination is checked and, if the browse node is missing, a request to create the browse node is generated. In some example embodiments, an entire set of missing browse nodes is generated based on recognition of a pattern of category and aspect combinations being requested.

Abstract: In various example embodiments, a system and method for a Target Language Engine are presented. The Target Language Engine augments a synonym list in a base dictionary of a target language with one or more historical search queries previously submitted to search one or more listings in listing data. The Target Language Engine identifies a compound word and a plurality of words present in the listing data that have a common meaning in the target language. Each word from the plurality of words is present in the compound word. The Target Language Engine causes a database to create an associative link between the portion of text and a word selected from at least one of the synonym list or the plurality of words.

Abstract: In various example embodiments, a system and method for retrieving database records are presented. A method includes receiving an access request for database records, identifying a primary database record corresponding to the access request, and selecting a set of secondary database records corresponding to a portion of the access request. The method determines a set of differentiating elements within records of the set of secondary database records. Each differentiating element distinguishes at least one record of the set of secondary database records from one or more of the primary database record and a portion of the records included in the set of secondary database records. The method generates a formatted search interface including a set of selectable interface icons representing one or more differentiating elements and causes presentation of the formatted search interface at a client device associated with the access request.

Abstract: In various example embodiments, a system and method for retrieving database records are presented. A method includes receiving an access request for database records, identifying a primary database record corresponding to the access request, and selecting a set of secondary database records corresponding to a portion of the access request. The method determines a set of differentiating elements within records of the set of secondary database records. Each differentiating element distinguishes at least one record of the set of secondary database records from one or more of the primary database record and a portion of the records included in the set of secondary database records. The method generates a formatted search interface including a set of selectable interface icons representing one or more differentiating elements and causes presentation of the formatted search interface at a client device associated with the access request.

Abstract: Systems, methods and media are provided for training a snippet extractor to create snippets based on information extracted from published descriptions. In one example, a computer-implemented method includes creating, based on a non-RNN (Recurrent Neural Network) extraction technique performed on the published descriptions, a plurality of base models, each base model including one or more sample description summaries; evaluating the base models using an evaluation technique; selecting an optimum base model; developing a classification model using RNN extraction, the classification model based on description summaries contained in the optimum base model; and using the classification model to train the snippet extractor by machine learning.

Abstract: In various example embodiments, a system and method for a Target Language Engine are presented. The Target Language Engine augments a synonym list in a base dictionary of a target language with one or more historical search queries previously submitted to search one or more listings in listing data. The Target Language Engine identifies a compound word and a plurality of words present in the listing data that have a common meaning in the target language. Each word from the plurality of words is present in the compound word. The Target Language Engine causes a database to create an associative link between the portion of text and a word selected from at least one of the synonym list or the plurality of words.

Abstract: In various example embodiments, a system and method for a Target Language Engine are presented. The Target Language Engine augments a synonym list in a base dictionary of a target language with one or more historical search queries previously submitted to search one or more listings in listing data. The Target Language Engine identifies a compound word and a plurality of words present in the listing data that have a common meaning in the target language. Each word from the plurality of words is present in the compound word. The Target Language Engine causes a database to create an associative link between the portion of text and a word selected from at least one of the synonym list or the plurality of words.

Abstract: Method for analyzing review content is described herein. The method starts by generating a helpfulness score for each review in the plurality of reviews. A first set of reviews is selected based on the helpfulness scores for each review in the plurality of review and a helpfulness score for each sentence in each review in the first set of reviews is generated. The review summary is then generated by selecting sentences in each review in the first set of reviews based on the helpfulness score of each sentence. The review summary includes the selected sentences. Other embodiments are also described.

Abstract: In various example embodiments, a system and method for a Target Language Engine are presented. The Target Language Engine augments a synonym list in a base dictionary of a target language with one or more historical search queries previously submitted to search one or more listings in listing data. The Target Language Engine identifies a compound word and a plurality of words present in the listing data that have a common meaning in the target language. Each word from the plurality of words is present in the compound word. The Target Language Engine causes a database to create an associative link between the portion of text and a word selected from at least one of the synonym list or the plurality of words.

Abstract: In various example embodiments, a system and method for evaluating and sorting product feedback is presented. In one example, a system includes a learning module to train a machine learning system on feedback from a plurality of users, the machine learning system configured to generate a quality rating for individual feedback, a feedback module to collect feedback for a specific product available from the online network based marketplace and apply the machine learning system to generate a quality rating for each feedback collected, and a sorting module to sort the feedback collected according to the quality ratings generated by the machine learning system.

Abstract: To improve the user experience, browse nodes for popular combinations of categories and aspects are desirable. When a particular combination of category and aspects is found to have been requested at a frequency that exceeds a threshold, the existence of a browse node for that combination is checked and, if the browse node is missing, a request to create the browse node is generated. In some example embodiments, an entire set of missing browse nodes is generated based on recognition of a pattern of category and aspect combinations being requested.

Abstract: A global aspect dictionary is created based on queries run on a search engine and responsive results to the query that are interacted with. The global aspect dictionary is used by a mapping server to look up aspects for queries received from users. A query received by the mapping server is modified by adding aspects from the dictionary to the query. in some example embodiments, tokens in the query corresponding to the added aspects are removed from the query. Results responsive to the modified query are transmitted to the user.

Abstract: In various example embodiments, a system and method for retrieving database records are presented. A method includes receiving an access request for database records, identifying a primary database record corresponding to the access request, and selecting a set of secondary database records corresponding to a portion of the access request. The method determines a set of differentiating elements within records of the set of secondary database records. Each differentiating element distinguishes at least one record of the set of secondary database records from one or more of the primary database record and a portion of the records included in the set of secondary database records. The method generates a formatted search interface including a set of selectable interface icons representing one or more differentiating elements and causes presentation of the formatted search interface at a client device associated with the access request.

Abstract: Systems, methods and media are provided for training a snippet extractor to create snippets based on information extracted from published descriptions. In one example, a computer-implemented method includes creating, based on a non-RNN (Recurrent Neural Network) extraction technique performed on the published descriptions, a plurality of base models, each base model including one or more sample description summaries; evaluating the base models using an evaluation technique; selecting an optimum base model; developing a classification model using RNN extraction, the classification model based on description summaries contained in the optimum base model; and using the classification model to train the snippet extractor by machine learning.