CN103593410B - System for search recommendation by means of replacing conceptual terms - Google Patents
System for search recommendation by means of replacing conceptual terms Download PDFInfo
- Publication number
- CN103593410B CN103593410B CN201310501114.XA CN201310501114A CN103593410B CN 103593410 B CN103593410 B CN 103593410B CN 201310501114 A CN201310501114 A CN 201310501114A CN 103593410 B CN103593410 B CN 103593410B
- Authority
- CN
- China
- Prior art keywords
- conceptual
- concept
- entity
- key word
- quality
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
Abstract
The invention provides a system for research recommendation by means of replacing conceptual terms. The system comprises an offline system and an online system; the offline system is used for parsing and identifying entity keywords contained in each historical record in log of search engines and creating indexes for the historical records according to categories of the entity keywords; the online system is used for receiving and parsing search engine queries submitted by users, identifying conceptual keywords in the search engine queries and searching historical queries according to weights, the historical queries are the most proximate to the given search engine queries and contain entity keywords with meaning of the conceptual keywords, and then the found queries are sorted and are returned to the users for secondary queries. The system has the advantages that the system is simple and direct, and massive data from the search engines are utilized; certain abstract conceptual keywords can be utilized when the users cannot provide accurate search terms; recommended search terms can be directly provided by the system, so that the user experience can be improved.
Description
Technical field
The present invention relates to natural language processing, searching engine field, in particular it relates to the control method of XX and corresponding
Control device.
Background technology
Jing retrievals find following coordinate indexing result:
Coordinate indexing result 1:
Application Number (patent):200580042218.2, title:Suggesting search engine keywords
Summary:Search engine receives the search inquiry with one or more keywords.Analysis is from the search inquiry
Document in result set, identifies one or more additional keywords for further splitting or separating initial result set.These are attached
Plus key word is presented to user, then user chooses whether to include or exclude the document for matching these additional keywords.With this
Mode, the number of documents that baseline results are concentrated is with relatively rapid and easily mode is reduced
The patent documentation is analyzed based on the result set of the search inquiry to user input, and is extracted and be can be used to
Then the key word for extracting is recommended user by the key word of segmentation result, determines it is to retain these key words to refer to by user
To document, still exclude the document that these key words are pointed to.Although it is succinct that this process seems comparison, for current
Big data epoch, user are difficult accurately to provide initial search inquiry, and in this case, the method cannot just ensure initially
Result set in the document that really needs comprising user, also cannot just ensure effectiveness.
Technical essential compares:
1. key word recommendation is carried out according to the result set of the search engine inquiry of user input, determine to include by user (or
Person excludes) result set of recommended keywords, and the search of the historical record and user input in the present invention using search engine is drawn
Holding up inquiry carries out the recommendation of whole inquiry.
2. direct key word recommendation is carried out for the result set of the search engine inquiry of user input, and the present invention attempts
Deeper level is obtained from the inquiry for semantically understanding user input, then carries out inquiry recommendation using semanteme.
Coordinate indexing result 2:
Application Number (patent):201010618555.4, title:The method and apparatus for recommending search keyword
Summary:This application discloses it is a kind of recommend search keyword method and apparatus, to solve in prior art to
During the recommendation search keyword of the user without clear and definite search intention, recommendation effect is not good, causes search engine server system resource
The problem of waste.Method includes:The search keyword of receives input;The search keyword for relatively receiving is not intended to word with setting
Sample word in set and the sample word in the intention set of words of setting;When comparative result is that the search keyword for receiving is included
The sample word that is not intended in set of words and when not including the sample word being intended in set of words, with the first predetermined way of recommendation to determine
Recommend the master mode of search keyword, with other ways of recommendation in addition to the first predetermined way of recommendation to determine search keyword
The strategy of supplementary mode, it is determined that recommend search keyword, wherein, the way of recommendation of the first predetermined way of recommendation for knowledge based storehouse
And/or the way of recommendation of dialogue-based dependency.
The patent documentation belongs to meaning using the search inquiry for being intended to word set and be not intended to word set to judge a user input
Figure inquiry is still not intended to inquiry, then according to the result for judging, using different strategies as main Generalization bounds.Work as search
When inquiry is judged as intent query, the patent is used based on conversation-based Generalization bounds.When search inquiry be judged as it is non-
During intent query, the patent is used based on the Generalization bounds in knowledge based storehouse.But it is intended to word set and to be not intended to word set itself non-
It is often limited, and need constantly to safeguard renewal, cost is larger;The knowledge base which uses simultaneously is also mainly with Alibaba Co
Based on ecommerce classification.
Technical essential compares:
1. using compound Generalization bounds, wherein, aid in for the recommendation of fuzzy query is also adopted by knowledge base, but
Its knowledge base main source is that the ecommerce of Alibaba Co is classified, and the knowledge base of the present invention adopts Probase or appoints
Anticipate a kind of probability hierarchical data base.
Coordinate indexing result 3:
Application Number (patent):201310165048.3, title:The recommendation method and search engine of search candidate word
Summary:The present invention proposes a kind of recommendation method and search engine of search candidate word, and wherein methods described includes:Search
The input information of rope engine server receiving user's input, and obtain the prefix information of input information;Using prefix information as rope
Draw the weight for obtaining multiple search candidate words and each search candidate word;Whether there is at least in judging multiple search candidate words
Two search candidate words belong to same subject;Belong to same subject if it is determined that having at least two and searching for candidate word, then retain
At least two search candidate words in one search candidate words weight it is constant, at least two search candidate words in other search
The weight of rope candidate word carries out drop power process;And be ranked up according to the weight of multiple search candidate words, after sequence
Search candidate word is provided to user.Method according to embodiments of the present invention, improves the multiformity and accuracy of search candidate word,
The search need of user is disclosure satisfy that, and algorithm is simple, it is easy to implemented, lift Consumer's Experience.
Prefix of the patent documentation mainly for the search inquiry of user input, scans for the recommendation inquired about, substantially
Say, equivalent to a kind of auto-complete function.In practical operation, this auto-complete function has a more serious defect, just
Be because of certain burst focus incident, and cause with regard to this burst focus numerous inquiries weight it is elevated together,
Its result is that search engine can recommend the extremely close search inquiry of numerous meanings.Although the patent is entered for this special screne
Optimization is gone, but has not still escaped for the dependence of accurate key word.
Technical essential compares
1. mainly for the polymerization drop power of similar recommended keywords, and the present invention tends to enter fuzzy key word of expressing the meaning
Row understands and recommends.
2. key word recommendation is scanned for mainly for prefix, and present invention is generally directed to the portion obscured in search keyword
Dividing carries out rewriting recommendation.
The content of the invention
For defect of the prior art, it is an object of the invention to provide a kind of scanned for by replacing conceptual word
The system of recommendation.The technical problem to be solved in the present invention be embodied in it is following some:
1) hierarchical knowledge base is introduced, the understanding of a conceptualization is carried out such that it is able to the search inquiry to user input,
Conceptual (ambiguous) key word therein is recognized.
2) as search engine is for having preferable performance based on the inquiry of key word, therefore the present invention is general by what is recognized
The property read key word replaces with more specifically entity (specific) key word, so as to obtain more preferable Search Results.
3) recommended using search engine logs.The user search queries of magnanimity are have recorded in the daily record of search engine,
Can therefrom filter out high-quality, the accurate search inquiry of result set, then these high-quality inquiries are recommended cannot provide precisely
The users of key word.This method both directly, can improve the Consumer's Experience of search engine again.
The system recommended is scanned for by replacing conceptual word according to what the present invention was provided, including off-line system and
Linear system is united, wherein:
Off-line system, for parsing the entity key word included in every historical record in identification search engine logs,
Then according to these classifications belonging to entity key word, are that these historical records set up index, so that on-line system is used;
On-line system, for receiving and parsing through the search engine inquiry submitted to by user, recognizes conceptual key therein
Word, then according to weight, finds closest with given search inquiry, and the entity comprising conceptual keyword sense
Then the inquiry for searching is ranked up by the historical query of key word, and returns to one recommendation after sequence of user
List, selects which to think the inquiry more pressed close to by user, carries out secondary inquiry.
Preferably, the off-line system includes entity abstraction module and concept aggregation module, wherein:
Entity abstraction module, for recognizing the entity key word included in every historical query, then will recognize
Then entity key word abstract gives the process of concept aggregation module to corresponding conceptual key word;
Concept aggregation module, for the historical query comprising same concept is aggregated to together, sets up index;For each
Bar historical query, entity abstraction module identify the entity key word and their corresponding concepts for wherein including, concept
Historical query comprising same concept is aggregated to together by aggregation module according to these concepts;One is set up with concept as major key
Index, give on-line system use.
Preferably, the on-line system includes conceptual analyses module, indexed search module and marking order module, wherein:
Conceptual analyses module, the conceptual key word in the search inquiry submitted to for identifying user;
Indexed search module, for the conceptual key word identified according to conceptual analyses module, travels through by off-line system
The index of generation, finds all historical querys comprising the entity key word consistent with the conceptual key word for identifying,
Inquire about these historical querys as Candidate Recommendation;
Marking order module, for the Candidate Recommendation inquiry marking found to all indexed search modules, and sorts, most
A part for sorted Candidate Recommendation list is returned to into user's selection afterwards.
Preferably, the marking is defined as distance, and which includes three parts:Semantic distance, literal distance and history
The quality of inquiry.
Preferably, the semantic distance is to inquire about original conceptual key word with the entity replaced for describing user
The typicality of key word, typicality is defined with equation below:
Wherein, Typicality (instance, concept) represents that for given concept one entity is general for this
The typical degree of thought, Freq (instance, concept) represent the frequency that an entity and a concept occur jointly, Freq
(concept) given frequency of the concept in corpus is represented, instance represents an entity, and concept represents one generally
Read;
And converted with equation below:
Wherein, SemDist (typ) represents semantic distance, and typ represents the value of typical case's degree, by Typicality
(instance, concept) formula is calculated, and e is the nature truth of a matter.
Preferably, the literal distance be for describing the other parts in addition to conceptual key word, user's inquiry with it is standby
Similarity between choosing inquiry.
Preferably, the quality of the historical query is for describing whether a historical query really has for a user
With, wherein, the quality of historical query is defined with equation below:
Wherein, Quality (query) represents the quality of given historical query, and ClickTime (query) represents user
Number of clicks in the result set of given historical query, Freq (query) represent the searched frequency of given historical query,
Query represents a given historical query;
And do such as down conversion to inquiring about quality:
QualityDist (quality)=e-quality
Wherein, QualityDist (quality) represents quality distance, and quality represents a mass value, by Quality
(query) formula is calculated.
Preferably, last composite score TotalScore of the marking is defined by equation below:
Wherein, SemDist represents semantic distance, and WordDist represents literal distance, QualityDist represent quality away from
From.
Compared with prior art, the present invention has following beneficial effect:
1) using the historical query in search engine logs as recommending data source, it is simple directly, and make use of and be derived from
The mass data of search engine.
2) when user cannot provide accurate search word, can be with some abstract concept key words so that
User can scan for action, without helpless.
3) directly give the search word of recommendation so that user can click on the search word of recommendation and carry out secondary inquiry, simply
It is convenient, improve Consumer's Experience.
Description of the drawings
Detailed description non-limiting example made with reference to the following drawings by reading, the further feature of the present invention,
Objects and advantages will become more apparent upon:
Fig. 1 is off-line system schematic diagram.
Fig. 2 is on-line system schematic diagram.
Fig. 3 is index structure schematic diagram.
Specific embodiment
With reference to specific embodiment, the present invention is described in detail.Following examples will be helpful to the technology of this area
Personnel further understand the present invention, but the invention is not limited in any way.It should be pointed out that the ordinary skill to this area
For personnel, without departing from the inventive concept of the premise, some deformations and improvement can also be made.These belong to the present invention
Protection domain.
According to present invention offer by the major function for replacing the system that conceptual word scans for recommendation it is
Using the historical record stored in search engine logs, index, Ran Houtong are set up for these historical records according to certain mode
Cross and matched with the search inquiry of user input, so as to by user may historical record interested recommend user.
The search commending system includes two subsystems, is off-line system and on-line system respectively.Off-line system is responsible for
Include in every historical record in parsing identification search engine logs entity key word (as " Obama ", " Nixon ",
" Watergate " etc.), then according to these classifications belonging to entity key word, are that these historical records set up index, for
On-line system is used.On-line system is responsible for receiving and parsing through the search engine inquiry submitted to by user, recognizes therein conceptual
Key word (such as " president ", " scandal " etc.), then according to weight, find it is closest with given search inquiry,
And then the inquiry for searching is ranked up by the historical query of the entity key word comprising conceptual keyword sense,
And one recommendation list after sequence of user is returned to, select which to think the inquiry more pressed close to by user, carry out secondary
Inquiry.
Off-line system mainly includes two modules:Entity abstraction module and concept aggregation module.As shown in Figure 1.
Entity abstraction module, entity key word which is included in being used to recognize every historical query (as " Obama ",
" Nixon ", " Watergate " etc.), then by the entity key word abstract for recognizing to corresponding conceptual key word (such as
" president ", " scandal " etc.), then give the process of concept aggregation module.Need exist for a stratification knowledge base
Auxiliary, knowledge base can provide the set of an entity key word, the set of a conceptual key word and a group object-
Corresponding relation between concept.
Concept aggregation module, which is used to be aggregated to together the historical query comprising same concept, sets up index.For every
One historical query, entity abstraction module can identify the entity key word for wherein including, and they are corresponding general
Read, concept aggregation module is sought to according to these concepts, the historical query comprising same concept is aggregated to together.Set up one
Index with concept as major key, gives on-line system use.
On-line system includes three modules:Conceptual analyses module, indexed search module and marking order module.Such as Fig. 2 institutes
Show.
Conceptual analyses module, its conceptual key word being used in the search inquiry of identifying user submission.For example, work as user
During input " president involved in scandals ", the module can identify two conceptual keys in inquiry
Word --- " president " and " scandal ".And following module is given by this information.
Indexed search module, which is used for according to the conceptual key word for identifying, travels through the index generated by off-line system,
All historical querys comprising the entity key word consistent with the conceptual key word for identifying are found, these history are looked into
Ask and inquire about as Candidate Recommendation, give next module and given a mark and sorted.
Marking order module, which is used for all Candidate Recommendation inquiry marking found, and sorts, and will finally sequence sequence
Candidate Recommendation list a part return to user selection.Three factors will be considered in the marking stage, first semantic factor,
I.e. in original query conceptual key word with recommend historical query in entity key word the degree of association;Which two is original query
The literal similarity between recommendation query;Its three be the historical query recommended quality, the present invention is main by inquiring about frequency
Rate (number of times being queried) and average hits (average to inquire about every time, hits of the user in result set).Return in list
Stage, needs be dynamically determined the threshold value (X) that returns inquiry and it is maximum return quantity (N), i.e., the recommendation query of all returns is divided
Number can not be more than X, while the recommendation query for returning is not more than N number of.
Implementation example 1:The foundation of historical query index
For on-line system, need, in the case where conceptual key word is given, to be quickly found out and given conceptual dependency
Historical query set, alternatively collects as recommendation query.Therefore, when historical query index is set up, it should with conceptual key word
As index key, whole index is organized.As shown in figure 3, all historical querys related to a certain specific concept are all concentrated
To in the query set with this concept as key assignments, the historical query such as comprising the president such as " Obama ", " Nixon " is all concentrated to
In query set with " president " as key assignments.Index is the set of the query set by all concepts as above.Especially
, concrete system realize in, for ask can fast access information, all concepts, historical query are all endowed unique identity
Indications (UID) are so as to quick access.
Implementation example 2:For the analysis of inquiry
For the analysis of inquiry refers mainly to recognize the conceptual key word or entity key word in inquiry.In off-line system
Entity abstraction module in, mainly the entity key word in historical query is identified;And in the general of on-line system
Read in analysis module, mainly to the inquiry of user input in conceptual key word be identified.The two is simply in identification
The data set different (the former is entity keyword set, and the latter is conceptual keyword set) for utilizing is identical in method.With bottom
Divide and will illustrate by taking online part as an example.
, from hierarchical knowledge base, level knowledge base is by a series of trees for the conceptual keyword set for using in the implementation
Shape structure is completing to the abstract of the world.Such as this logic chain:“market”→“emerging market”→“China、
India…”." China " and " India " these belong to entity key word, to they carry out one it is abstract compared with shallow hierarchy, can
It is to be conceptualized as " emerging market " (emerging market), further abstract, " market " (city can be conceptualized as
).The information source of this knowledge base can be captured from network by network information acquiring technology.
There is conceptual keyword set, it is possible to the conceptual key word in user's inquiry is identified, the present invention is adopted
Mode is longest match principle, i.e., when " emerging market " is identified, just no longer recognize " market ".This is main
Because longer concept, more accurate concept is generally meant that, so as to cover more accurate entity keyword set, obtained then
Obtain more accurate recommendation.
Implementation example 3:The method of marking sequence
By analysis, it can be appreciated that the fuzzy query of user's submission can include more than one conceptual key
Word, by traveling through index, can find comprising all historical querys for recognizing the corresponding entity of conceptual key word, and these are gone through
Alternately recommendation query will be scored sequence for history inquiry.Marking in the present invention is defined as distance (more near better), and which includes
Three parts:The quality of semantic distance, literal distance and historical query.
Semantic distance is the allusion quotation that original conceptual key word and the entity key word replaced are inquired about for describing user
Type.For example, " president " is replaced with " Barack Obama " just than being replaced with " Bill " typical case, because
In known, " Barack Obama " refers almost exclusively to incumbent US President, and " Bill " is then very common U.S.'s name.Tool
Body, can be defined with equation below with regard to typicality:
Wherein, Typicality (instance, concept) represents that for given concept one entity is general for this
The typical degree of thought, Freq (instance, concept) represent the frequency that an entity and a concept occur jointly,
Freq (concept) represents given frequency of the concept in corpus, and instance represents an entity, and concept represents one
Individual concept.
According to formula, when an entity is more typical for a concept, this numerical value is bigger, to accord with this numerical value
The concept of " distance " is closed, therefore we need to be converted with equation below:
Wherein, SemDist (typ) represents semantic distance, and typ represents the value of typical case's degree, by Typicality
(instance, concept) formula is calculated.
Literal distance (WordDist) be for describing the other parts in addition to conceptual key word, user's inquiry with it is alternative
Similarity between inquiry.Be employed herein " between character string, changing distance " for commonly using in Computer Subject as it is literal away from
From.
Whether the quality of historical query is genuine useful for a user for describing a historical query.Based on useful
Inquiry, be more possible to by user click on it is multiple it is assumed that in the present invention with equation below define historical query quality:
Wherein, Quality (query) represents the quality of given historical query, and ClickTime (query) represents user
Number of clicks in the result set of given historical query, Freq (query) represent the searched frequency of given historical query,
Query represents a given historical query.
It is similar with semantic distance, in order to meet the definition of " distance ", need to do such as down conversion to inquiring about quality:
QualityDist (quality)=e-quality
Wherein, QualityDist (quality) represents quality distance, and quality represents a mass value, by Quality
(query) formula is calculated.
Last composite score TotalScore is defined by equation below:
Wherein, SemDist represents semantic distance, and WordDist represents literal distance, QualityDist represent quality away from
From.
I.e. semantic distance and literal distance linear and after rounding up, adds quality distance.This calculation is caused
The distance for calculating (rounds) degree of association that ensure that between original query and recommendation query in certain granularity, in turn ensure that by
The historical query of recommendation has enough quality.
Above the specific embodiment of the present invention is described.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can make various modifications or modification within the scope of the claims, this not shadow
Ring the flesh and blood of the present invention.
Claims (5)
1. it is a kind of to scan for the system recommended by replacing conceptual word, it is characterised in that including off-line system and online
System, wherein:
Off-line system, for parsing the entity key word included in every historical record in identification search engine logs, then
According to the classification belonging to these entity key words, it is that these historical records set up index, so that on-line system is used;
On-line system, for receiving and parsing through the search engine inquiry submitted to by user, recognizes conceptual key word therein, so
Afterwards according to weight, find that closest with given search inquiry, and the entity comprising conceptual keyword sense is crucial
Then the inquiry for searching is ranked up by the historical query of word, and returns to one recommendation list after sequence of user,
Select which to think the inquiry more pressed close to by user, carry out secondary inquiry;
The on-line system includes conceptual analyses module, indexed search module and marking order module, wherein:
Conceptual analyses module, the conceptual key word in the search inquiry submitted to for identifying user;
Indexed search module, for the conceptual key word identified according to conceptual analyses module, travels through and is generated by off-line system
Index, find all historical querys comprising the entity key word consistent with the conceptual key word for identifying, by this
A little historical querys are inquired about as Candidate Recommendation;
Marking order module, for the Candidate Recommendation inquiry marking found to all indexed search modules, and sorts, finally will
A part for sorted Candidate Recommendation list returns to user's selection;
The marking is defined as distance, and which includes three parts:The quality of semantic distance, literal distance and historical query;
The semantic distance is the allusion quotation that original conceptual key word and the entity key word replaced are inquired about for describing user
Type, typicality is defined with equation below:
Wherein, Typicality (instance, concept) represents that for given concept one entity is for this concept
Typical degree, Freq (instance, concept) represent the frequency that an entity and a concept occur jointly, Freq
(concept) given frequency of the concept in corpus is represented, instance represents an entity, and concept represents one generally
Read;
And converted with equation below:
Wherein, SemDist (typ) represents semantic distance, and typ represents the value of typical case's degree, by Typicality
(instance, concept) formula is calculated, and e is the nature truth of a matter.
2. it is according to claim 1 to scan for the system recommended by replacing conceptual word, it is characterised in that described
Off-line system includes entity abstraction module and concept aggregation module, wherein:
Entity abstraction module, for recognizing the entity key word included in every historical query, then by the entity for recognizing
Property key word abstract to corresponding conceptual key word, then give concept aggregation module process;
Concept aggregation module, for the historical query comprising same concept is aggregated to together, sets up index;Each is gone through
History is inquired about, and entity abstraction module identifies the entity key word and their corresponding concepts for wherein including, concept polymerization
Historical query comprising same concept is aggregated to together by module according to these concepts;Set up a rope with concept as major key
Draw, give on-line system use.
3. it is according to claim 1 to scan for the system recommended by replacing conceptual word, it is characterised in that described
Literal distance is for describing the other parts in addition to conceptual key word, the similarity between user's inquiry and alternative inquiry.
4. it is according to claim 1 to scan for the system recommended by replacing conceptual word, it is characterised in that described
The quality of historical query be it is whether genuine useful for a user for describing a historical query, wherein, use equation below
Define the quality of historical query:
Wherein, Quality (query) represents the quality of given historical query, and ClickTime (query) represents that user is giving
Determine the number of clicks in the result set of historical query, Freq (query) represents the searched frequency of given historical query, query
Represent a given historical query;
And do such as down conversion to inquiring about quality:
QualityDist (quality)=e-quality
Wherein, QualityDist (quality) represents quality distance, and quality represents a mass value, by Quality
(query) formula is calculated, and e is the nature truth of a matter.
5. it is according to claim 1 to scan for the system recommended by replacing conceptual word, it is characterised in that described
Last composite score TotalScore of marking is defined by equation below:
Wherein, SemDist represents semantic distance, and WordDist represents literal distance, and QualityDist represents quality distance.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310501114.XA CN103593410B (en) | 2013-10-22 | 2013-10-22 | System for search recommendation by means of replacing conceptual terms |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310501114.XA CN103593410B (en) | 2013-10-22 | 2013-10-22 | System for search recommendation by means of replacing conceptual terms |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103593410A CN103593410A (en) | 2014-02-19 |
CN103593410B true CN103593410B (en) | 2017-04-12 |
Family
ID=50083551
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310501114.XA Expired - Fee Related CN103593410B (en) | 2013-10-22 | 2013-10-22 | System for search recommendation by means of replacing conceptual terms |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103593410B (en) |
Families Citing this family (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104063418A (en) * | 2014-03-17 | 2014-09-24 | 百度在线网络技术(北京)有限公司 | Search recommendation method and device |
CN104199810A (en) * | 2014-08-29 | 2014-12-10 | 科大讯飞股份有限公司 | Intelligent service method and system based on natural language interaction |
CN104503975B (en) * | 2014-11-20 | 2021-08-06 | 百度在线网络技术(北京)有限公司 | Recommendation card customization method and device |
CN104484339B (en) * | 2014-11-21 | 2018-01-26 | 百度在线网络技术(北京)有限公司 | A kind of related entities recommend method and system |
CN104933100B (en) * | 2015-05-28 | 2018-05-04 | 北京奇艺世纪科技有限公司 | keyword recommendation method and device |
US20160371340A1 (en) * | 2015-06-19 | 2016-12-22 | Lenovo (Singapore) Pte. Ltd. | Modifying search results based on context characteristics |
CN104991955B (en) * | 2015-07-17 | 2018-06-12 | 安徽科大讯飞医疗信息技术有限公司 | Method and system for automatically constructing template library |
CN105868249A (en) * | 2015-12-15 | 2016-08-17 | 乐视网信息技术(北京)股份有限公司 | Data query control method and device |
CN105868255A (en) * | 2015-12-25 | 2016-08-17 | 乐视网信息技术(北京)股份有限公司 | Query recommendation method and apparatus |
WO2018112696A1 (en) * | 2016-12-19 | 2018-06-28 | 深圳大学 | Content pushing method and content pushing system |
CN106649750B (en) * | 2016-12-26 | 2021-02-05 | 三六零科技集团有限公司 | Searching method and device for multi-meaning term entry |
CN108460039A (en) * | 2017-02-20 | 2018-08-28 | 微软技术许可有限责任公司 | Recommendation is provided |
CN107679186B (en) * | 2017-09-30 | 2021-12-21 | 北京奇虎科技有限公司 | Method and device for searching entity based on entity library |
CN110807094A (en) * | 2018-07-20 | 2020-02-18 | 林威伶 | Big data analysis, prediction and data visualization system and device for legal document |
CN109558479B (en) * | 2018-11-29 | 2022-12-02 | 出门问问创新科技有限公司 | Rule matching method, device, equipment and storage medium |
CN110442696B (en) * | 2019-08-05 | 2022-07-08 | 北京百度网讯科技有限公司 | Query processing method and device |
WO2021136009A1 (en) * | 2019-12-31 | 2021-07-08 | 阿里巴巴集团控股有限公司 | Search information processing method and apparatus, and electronic device |
CN111475725B (en) * | 2020-04-01 | 2023-11-07 | 百度在线网络技术(北京)有限公司 | Method, apparatus, device and computer readable storage medium for searching content |
CN111782935A (en) * | 2020-05-12 | 2020-10-16 | 北京三快在线科技有限公司 | Information recommendation method and device, electronic equipment and storage medium |
CN112784171B (en) * | 2021-01-21 | 2022-11-04 | 重庆邮电大学 | Movie recommendation method based on context typicality |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102087699A (en) * | 2011-03-03 | 2011-06-08 | 北京天地融科技有限公司 | Information transmission method and system, bar code display device and reading device |
CN102591969A (en) * | 2011-12-31 | 2012-07-18 | 北京百度网讯科技有限公司 | Method for providing search results based on historical behaviors of user and sever therefor |
CN103218362A (en) * | 2012-01-19 | 2013-07-24 | 中兴通讯股份有限公司 | Method and system for constructing domain ontology |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8010545B2 (en) * | 2008-08-28 | 2011-08-30 | Palo Alto Research Center Incorporated | System and method for providing a topic-directed search |
-
2013
- 2013-10-22 CN CN201310501114.XA patent/CN103593410B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102087699A (en) * | 2011-03-03 | 2011-06-08 | 北京天地融科技有限公司 | Information transmission method and system, bar code display device and reading device |
CN102591969A (en) * | 2011-12-31 | 2012-07-18 | 北京百度网讯科技有限公司 | Method for providing search results based on historical behaviors of user and sever therefor |
CN103218362A (en) * | 2012-01-19 | 2013-07-24 | 中兴通讯股份有限公司 | Method and system for constructing domain ontology |
Also Published As
Publication number | Publication date |
---|---|
CN103593410A (en) | 2014-02-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103593410B (en) | System for search recommendation by means of replacing conceptual terms | |
Ritze et al. | Matching web tables to dbpedia-a feature utility study | |
US8676815B2 (en) | Suffix tree similarity measure for document clustering | |
US20030004942A1 (en) | Method and apparatus of metadata generation | |
US20110208776A1 (en) | Method and apparatus of semantic technological approach based on semantic relation in context and storage media having program source thereof | |
CN107992633A (en) | Electronic document automatic classification method and system based on keyword feature | |
Chaudhuri et al. | Exploiting web search to generate synonyms for entities | |
CN113312474A (en) | Similar case intelligent retrieval system of legal documents based on deep learning | |
CN112270188B (en) | Questioning type analysis path recommendation method, system and storage medium | |
CN115563313A (en) | Knowledge graph-based document book semantic retrieval system | |
US20120130999A1 (en) | Method and Apparatus for Searching Electronic Documents | |
CN115618014A (en) | Standard document analysis management system and method applying big data technology | |
CN112685440B (en) | Structural query information expression method for marking search semantic role | |
Elfayoumy et al. | A survey of unstructured text summarization techniques | |
CN115953041A (en) | Construction scheme and system of operator policy system | |
Fan et al. | Stop words for processing software engineering documents: Do they matter? | |
Guilly et al. | Sql query completion for data exploration | |
CN111259145B (en) | Text retrieval classification method, system and storage medium based on information data | |
Nagaraj et al. | A novel semantic level text classification by combining NLP and Thesaurus concepts | |
Zhang et al. | An adaptive method for organization name disambiguation with feature reinforcing | |
Perez-Guadarramas et al. | Analysis of OWA operators for automatic keyphrase extraction in a semantic context | |
Ajitha et al. | EFFECTIVE FEATURE EXTRACTION FOR DOCUMENT CLUSTERING TO ENHANCE SEARCH ENGINE USING XML. | |
Schadd et al. | Improving ontology matchers utilizing linguistic ontologies: an information retrieval approach | |
Faisal et al. | Contextual Word Embedding based Clustering for Extractive Summarization | |
Hemayati et al. | Categorizing search results using wordnet and Wikipedia |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170412 Termination date: 20191022 |
|
CF01 | Termination of patent right due to non-payment of annual fee |