FIELD: information technologies.
SUBSTANCE: invention relates to search results ranking engines, particularly to method of optimising search results ranking engines. Method of optimising search results ranking obtained from search results ranking engine, the first set of pairs of request-document, each of which has corresponding vector of post review parameters. Weight vector is generated with number of weight values corresponding to number of parameters of post review in each of vectors post review parameters of first set. Target function is generated using weighting vector and post review vectors parameters of first set. Weight vector weight values are optimised using first set of pairs of request-document for producing optimised target function using metrics efficiency associated with target function. Search results ranking engine is optimised using optimised target function and use optimised search results ranking engine for search results ranking.
EFFECT: high relevancy of search results, faster user search, reduced traffic and load on search server and web resources servers.
17 cl, 4 dwg
Title | Year | Author | Number |
---|---|---|---|
METHOD AND SERVER FOR TRAINING MACHINE LEARNING ALGORITHM IN OBJECT RANKING | 2020 |
|
RU2782502C1 |
METHOD AND SERVER FOR GENERATING META-ATTRIBUTE FOR RANGING DOCUMENTS | 2018 |
|
RU2721159C1 |
METHOD AND SYSTEM OF SEARCH QUERY PROCESSING | 2015 |
|
RU2640639C2 |
SEARCH INDEX CONSTRUCTION METHOD AND SYSTEM USING MACHINE LEARNING ALGORITHM | 2018 |
|
RU2720954C1 |
METHOD AND SYSTEM FOR DETECTION OF ABNORMAL RANKING | 2019 |
|
RU2776034C2 |
METHOD AND SYSTEM FOR CREATING ANNOTATION VECTORS FOR DOCUMENT | 2017 |
|
RU2720074C2 |
SYSTEM AND METHOD OF FORMING TRAINING SET FOR MACHINE LEARNING ALGORITHM | 2018 |
|
RU2744029C1 |
METHOD AND SYSTEM OF SELECTION FOR RANKING SEARCH RESULTS USING MACHINE LEARNING ALGORITHM | 2018 |
|
RU2731658C2 |
METHOD AND SYSTEM FOR GENERATING FEATURE FOR RANGING DOCUMENT | 2018 |
|
RU2733481C2 |
METHOD AND SYSTEM FOR EXPANDING SEARCH QUERIES IN ORDER TO RANK SEARCH RESULTS | 2018 |
|
RU2720905C2 |
Authors
Dates
2017-01-25—Published
2014-06-30—Filed