[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-123400-en":3,"doc-seo-123400-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},123400,13056703019662,"Evangeline","https://ap-avatar.wpscdn.com/avatar/be000253a8e92610077?_k=1778726343310543188",8,"Research & Report","Machine Learning Driven Ranking to Control Degree of Personalization in Search Results","Personalization in query-based search retrieval creates a core tension between query relevance and user affinity. This disclosure presents techniques for training a machine learning model to personalize search-result ranking in a controllable way. With user permission, training data is built from query logs by detecting manual refinements followed by long clicks, forming triplets of original query, refined query, and click. A model is trained with Softmax listwise loss to rank the refined-query click above clicks from the original query, and the personalization level versus relevance is adjusted by tuning the Softmax temperature.","Technical Disclosure Commons  \nDefensive Publications Series  \n06 Aug 2025  \nMachine Learning Driven Ranking to Control Degree of Personalization in Search Results  \nEric Lin  \nDevora Berlowitz  \nOlive Guo  \nFollow this and additional works at: [https://www.tdcommons.org/dpubs_series](https://www.tdcommons.org/dpubs_series)  \nRecommended Citation  \nLin, Eric; Berlowitz, Devora; and Guo, Olive, \"Machine Learning Driven Ranking to Control Degree of Personalization in Search Results\", Technical Disclosure Commons,(August 06, 2025)  \n[https://www.tdcommons.org/dpubs_series/8435](https://www.tdcommons.org/dpubs_series/8435)  \nThis work is licensed under a Creative Commons Attribution 4.0 License.  \nThis Article is brought to you for free and open access by Technical Disclosure Commons. It has been accepted for inclusion in Defensive Publications Series by an authorized administrator of Technical Disclosure Commons.  \nMachine Learning Driven Ranking to Control Degree of Personalization in Search  \nResults  \nABSTRACT  \nAn important problem in introducing personalization to query-based search retrieval and ranking is how to balance the relevance of results to the query with the affinity of the results to the user as oftentimes there is a trade-off between the two. This disclosure describes techniques to train a machine learning model to personalize ranking of search results in a controllable manner. With user permission, training data is constructed by analyzing query logs to identify manual query refinements followed by \"long clicks,\" forming triplets of {original query, manually refined query, click}. These reflect personalized search results that the user found of high quality. Clicks from the original query are also obtained, reflecting other results for the  \nquery that were viewed. A machine learning model is trained using Softmax listwise loss to  \nrank the true click from the manually refined query above clicks from the original query. The  \ndegree of personalization versus relevance can be controlled by tuning the temperature of the  \nSoftmax loss. The described techniques can be used in various search contexts.  \nKEYWORDS  \n● Personalized search results  \n● Search personalization  \n● Personalized rank  \n● Search engine  \n● Query log  \n● Long click  \n● Softmax loss  \n● Query triplet  \nPublished by Technical Disclosure Commons, 2025 2  \nBACKGROUND  \nAn important problem in introducing personalization to query-based search retrieval and ranking is how to balance the relevance of results to the query with the affinity of the results to the user as oftentimes there is a trade-off between the two. When a user has turned off personalization or does not provide permission to use user data to determine user interests for ranking purposes, default search results can be provided. However, this can be unsatisfactory for users that have provided permissions and that have an expectation that the search results take into account their interests, e.g., based on past history of queries and corresponding results, user interests determined with user permission, other information such as user location or  \ndemographic data, etc.  \nFor example, consider a user that often reads about baseball on a particular website that hosts community forums, e.g., discussions between various users that post on the topic. If the user issues the query \"baseball\", the search engine can return generic information about  \nbaseball (e.g., a national league, local league) or can return discussion threads related to  \nbaseball on the particular website in the search results based on their user history. When the  \nsame user issues the query \"basketball\", the search engine can either return generic information about basketball or can return discussion threads related to basketball on the particular website  \nsince the user historically has like sports discussion threads at the website.  \nDESCRIPTION  \nTo ensure search quality and enhance user satisfaction, search engine","cbCaikAfW5MBjJiz","https://ap.wps.com/l/cbCaikAfW5MBjJiz","pdf",221573,1,7,"English","en",105,"# Abstract\n# Background\n# Description\n## Training Data Construction\n## Model Training and Loss Function\n## Control of Personalization vs. Relevance","[{\"question\":\"What problem does the disclosure address in personalized search ranking?\",\"answer\":\"It addresses how to balance the relevance of results to the query with the affinity of results to the user, which can otherwise trade off against each other.\"},{\"question\":\"How is training data constructed for the personalized ranking model?\",\"answer\":\"With user permission, query logs are analyzed to find manual query refinements followed by long clicks, producing triplets of original query, refined query, and click.\"},{\"question\":\"How is the degree of personalization controlled during training?\",\"answer\":\"The model uses Softmax listwise loss, and the personalization versus relevance balance is controlled by tuning the Softmax temperature.\"}]","Machine Learning Driven Ranking to Control Degree of Personalization in Search Results | PDF",1785816290,18,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"machine-learning-driven-ranking-to-control-degree-of-personalization-in-search-results","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-driven-ranking-to-control-degree-of-personalization-in-search-results/123400/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the disclosure address in personalized search ranking?","Question",{"text":75,"@type":76},"It addresses how to balance the relevance of results to the query with the affinity of results to the user, which can otherwise trade off against each other.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is training data constructed for the personalized ranking model?",{"text":80,"@type":76},"With user permission, query logs are analyzed to find manual query refinements followed by long clicks, producing triplets of original query, refined query, and click.",{"name":82,"@type":73,"acceptedAnswer":83},"How is the degree of personalization controlled during training?",{"text":84,"@type":76},"The model uses Softmax listwise loss, and the personalization versus relevance balance is controlled by tuning the Softmax temperature.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]