Disambiguating unrecognized abbreviations in search queries using machine learning
US-2024070178-A1 · Feb 29, 2024 · US
US9619564B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-9619564-B2 |
| Application number | US-201414339715-A |
| Country | US |
| Kind code | B2 |
| Filing date | Jul 24, 2014 |
| Priority date | Jul 29, 2013 |
| Publication date | Apr 11, 2017 |
| Grant date | Apr 11, 2017 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
Embodiments of the present application relate to a method, a system, and a computer program product for providing recommended terms. A method for providing recommended terms is provided. The method includes determining a search query input by a user and an input time point when the search query was input, determining whether the input time point falls within a time range preset within a current time period, if the input time point falls within the time range preset within the current time period, determining whether the search query is present in a word bank preset for the time range, and if the search query is present in the word bank, providing the user with a special event recommended term preset for the time range.
Opening claim text (preview).
What is claimed is: 1. A method for providing recommended terms, comprising: determining a search query input by a user and an input time point when the search query was input; determining whether the input time point falls within a time range preset within a current time period; in the event that the input time point falls within the time range preset within the current time period, determining whether the search query input is present in a word bank preset for the time range; in the event that the search query input is present in the word bank, providing the user with a special event recommended term preset for the time range; in response to the user selecting the special event recommended term, jumping to a special event page preset for the time range; and presetting the word bank for the time range, comprising: dividing a set interval into a plurality of sub-intervals, the set interval corresponding to a designated time point within a previous time range; determining, for an unfinalized search query of unfinalized search queries within the plurality of sub-intervals, whether past search frequencies of the unfinalized search queries satisfy a normal distribution; and in the event that the past search frequencies of the unfinalized search queries satisfy the normal distribution, adding the selected unfinalized search query to the word bank preset for the time range. 2. A method for providing recommended terms, comprising: determining a search query input by a user and an input time point when the search query was input; determining whether the input time point falls within a time range preset within a current time period; in the event that the input time point falls within the time range preset within the current time period, determining whether the search query input is present in a word bank preset for the time range; in the event that the search query input is present in the word bank, providing the user with a special event recommended term preset for the time range; in response to the user selecting the special event recommended term, jumping to a special event page preset for the time range; and presetting the word bank for the time range, comprising: determining in advance within a previous time period a corresponding time point corresponding to a designated time point within the time range; dividing a first set interval prior to the corresponding time point within the previous time period into a plurality of first sub-intervals; for a first sub-interval, determining search queries searched within the plurality of first sub-intervals as unfinalized search queries, and selecting a first set quantity of unfinalized search queries from the unfinalized search queries in the plurality of first sub-intervals sorted in order of greater to lesser search frequency; determining a second set interval, the corresponding time point being an average time point of the second set interval, the second set interval relating to the corresponding time point; dividing the second set interval equally into a plurality of second sub-intervals; for a selected unfinalized search query, determining past search frequencies of unfinalized search queries of the selected first set quantity of the unfinalized search queries within the plurality of second sub-intervals, and determining whether the past search frequencies of the unfinalized search queries in the plurality of second sub-intervals satisfy a normal distribution; in the event that the past search frequencies of the unfinalized search query in the plurality of second sub-intervals satisfy the normal distribution, adding the unfinalized search query to the word bank preset for the time range; and in the event that the past search frequencies of the unfinalized search query in the plurality of second sub-intervals do not satisfy the normal distribution, omitting adding the unfinalized search query to the word bank preset for the time range. 3. The method as described in claim 2 , wherein the determining of whether the past search frequencies of the unfinalized search query in the second sub-intervals satisfy the normal distribution comprises: in the event that a quantity of divided second sub-intervals is not greater than a second set quantity of unfinalized search queries, determining whether the past search frequencies of the unfinalized search query in the second sub-intervals satisfy a normal distribution based on a W-test; and in the event that the quantity of divided second sub-intervals is greater than the second set quantity of unfinalized search queries, determining whether the past search frequencies of the unfinalized search query in the second sub-intervals satisfy a normal distribution based on a D-test. 4. The method as described in claim 2 , wherein the adding of the unfinalized search query to the word bank preset for the time range comprises: extracting a head of the unfinalized search query; and adding the extracted head to the word bank preset for the time range. 5. The method as described in claim 2 , wherein prior to the adding of the unfinalized search query to the word bank preset for the time range: determining that a mean value of a click rate for all information located with the unfinalized search query is not less than a set click rate. 6. The method as described in claim 2 , wherein the determining of whether the search query input is present in the word bank for the time range comprises: extracting a head from the search query input by the user; determining whether the head extracted from the search query input by the user is present in the word bank; in the event that the head extracted from the search query input by the user is present in the word bank, determining that the search query input by the user is present in the word bank; and in the event that the head extracted from the search query input by the user is not present in the word bank, determining that the search query input by the user is not present in the word bank. 7. The method as described in claim 2 , further comprising: presetting the special event page for the time range, comprising: predetermining information obtained by searching with each unfinalized search query added to the word bank; and adding the information to the special event page preset for the time range. 8. The method as described in claim 2 , further comprising: in the event that the search query input is not present in the word bank, providing the user with a recommended term associated with the search query input that the user has inputted, the recommended term associated with the search query input being different than the special event recommended term. 9. A system for providing recommended terms, comprising: at least one processor configured to: determine a search query input by a user and an input time point when the search query was input; determine whether the input time point falls within a time range preset within a current time period; in the event that the input time point falls within the time range preset within the current time period, determine whether the search query input is present in a word bank preset for the time range; in the event that the search query input is present in the word bank, provide the user with a special event recommended term preset for the time range; in response to the user selecting the special event recommended term, jump to a special event page preset for the time range; and preset the word bank for the time range, comprising: divide a set interval into a plurality of sub-intervals, the set interval corresponding to a designated time point within a previous time range; determine, for an unfinalized search query of unfinalized
Dictionaries · CPC title
Lexical analysis, e.g. tokenisation or collocates · CPC title
using system suggestions (G06F16/3325 takes precedence) · CPC title
Search customisation based on user profiles and personalisation · CPC title
Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.