CN105808688A - Complementation retrieval method and device based on artificial intelligence - Google Patents
Complementation retrieval method and device based on artificial intelligence Download PDFInfo
- Publication number
- CN105808688A CN105808688A CN201610119571.6A CN201610119571A CN105808688A CN 105808688 A CN105808688 A CN 105808688A CN 201610119571 A CN201610119571 A CN 201610119571A CN 105808688 A CN105808688 A CN 105808688A
- Authority
- CN
- China
- Prior art keywords
- query statement
- completion
- phrase
- statement
- user
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention puts forward a complementation retrieval method and device based on artificial intelligence. The method comprises the following steps: receiving a first query statement input by a user at present, and judging whether the first query statement meets a complementation condition or not according to a preset complementation retrieval strategy; if the first query statement meets the complementation condition, judging whether a second query statement previously retrieved by the user is associated with the first query statement or not; if the second query statement is associated with the first query statement, carrying out complementation on the first query statement according to the second query statement to generate a third query statement; and retrieving the third query statement, and feeding back a retrieval result to the user. Therefore, through a multi-round retrieval way, the interaction efficiency of a search engine and the user is improved, and time for the user to input the query statement is saved.
Description
Technical field
The application relates to artificial intelligence's technical field of information retrieval, particularly relates to a kind of completion search method based on artificial intelligence and device.
Background technology
Artificial intelligence (ArtificialIntelligence), english abbreviation is AI.It is that research, exploitation are for simulating, extend and extend new science of technology of the theory of intelligence of people, method, technology and application system.Artificial intelligence is a branch of computer science, the essence of intelligence is understood in its attempt, and produce a kind of new intelligent machine can made a response in the way of human intelligence is similar, the research in this field includes robot, speech recognition, image recognition, natural language processing and specialist system etc..Wherein, the most important aspect of artificial intelligence is exactly speech recognition technology.
Along with popularizing of the Internet, the information on the Internet is also more and more abundanter, and present people can obtain, by search engine, the information oneself wanted easily.
The mobile Internet being terminal with smart mobile phone in recent years is expanded rapidly, and the application relying on intelligent mobile phone terminal in a large number is also arisen at the historic moment.But smart mobile phone has compared its particularity with computer, for instance: the screen size of smart mobile phone and input through keyboard than computer little a lot；From interaction time, it is mutual that smart phone user is more likely to fragment type, as etc. the gap of public transport.As can be seen here, smart mobile phone is utilized to scan for mutual efficiency very low.
Summary of the invention
One of technical problem that the application is intended to solve in correlation technique at least to a certain extent.
For this, first purpose of the application is in that to propose a kind of completion search method based on artificial intelligence, the method achieves and improves search engine and user mutual efficiency by taking turns retrieval mode more, saves the time of user input query statement.
Second purpose of the application is in that proposing a kind of completion based on artificial intelligence retrieves device.
For reaching above-mentioned purpose, the application first aspect embodiment proposes a kind of completion search method based on artificial intelligence, including: receive the first query statement that user is currently entered, judge whether described first query statement meets completion condition according to default completion search strategy；If described first query statement meets completion condition, then judge that whether the second query statement of the upper retrieval of described user is relevant to described first query statement；If described second query statement is relevant to described first query statement, then according to described second query statement, described first query statement is carried out completion and generate the 3rd query statement；Described 3rd query statement is retrieved, and retrieves result to described user feedback.
According to default completion search strategy, the completion search method based on artificial intelligence of the embodiment of the present application, by receiving the first query statement that user is currently entered, judges whether described first query statement meets completion condition；If described first query statement meets completion condition, then judge that whether the second query statement of the upper retrieval of described user is relevant to described first query statement；If described second query statement is relevant to described first query statement, then according to described second query statement, described first query statement is carried out completion and generate the 3rd query statement；Described 3rd query statement is retrieved, and retrieves result to described user feedback.Hereby it is achieved that improved search engines and the mutual efficiency of user by many wheel retrieval modes, save the time of user input query statement.
For reaching above-mentioned purpose, the application second aspect embodiment proposes a kind of completion based on artificial intelligence and retrieves device, including: receiver module, for receiving the first query statement that user is currently entered；According to the completion search strategy preset, first judge module, for judging whether described first query statement meets completion condition；Second judge module, if meeting completion condition for described first query statement, then judges that whether the second query statement of the upper retrieval of described user is relevant to described first query statement；Completion module, if relevant to described first query statement for described second query statement, then carry out completion according to described second query statement to described first query statement and generates the 3rd query statement；Retrieval module, for described 3rd query statement is retrieved, and retrieves result to described user feedback.
Device is retrieved in the completion based on artificial intelligence of the embodiment of the present application, by receiving the first query statement that user is currently entered, judges whether described first query statement meets completion condition according to default completion search strategy；If described first query statement meets completion condition, then judge that whether the second query statement of the upper retrieval of described user is relevant to described first query statement；If described second query statement is relevant to described first query statement, then according to described second query statement, described first query statement is carried out completion and generate the 3rd query statement；Described 3rd query statement is retrieved, and retrieves result to described user feedback.Hereby it is achieved that improved search engines and the mutual efficiency of user by many wheel retrieval modes, save the time of user input query statement.
Accompanying drawing explanation
The present invention above-mentioned and/or that add aspect and advantage will be apparent from easy to understand from the following description of the accompanying drawings of embodiments, wherein:
Fig. 1 is the flow chart of the completion search method based on artificial intelligence of one embodiment of the application；
Fig. 2 is the flow chart of the completion search method based on artificial intelligence of another embodiment of the application；
Fig. 3 is based on entity attribute knowledge base and takes turns the system-wide technology frame chart of benefit more；
Fig. 4 is the schematic diagram being realized many wheel search by completion retrieval mode；
Fig. 5 is an exemplary plot of traditional single-wheel search engine；
Fig. 6 is that the schematic diagram supplementing retrieval taken turns by search engine more；
Fig. 7 is the structural representation of the completion retrieval device based on artificial intelligence of one embodiment of the application.
Detailed description of the invention
Being described below in detail embodiments herein, the example of described embodiment is shown in the drawings, and wherein same or similar label represents same or similar element or has the element of same or like function from start to finish.The embodiment described below with reference to accompanying drawing is illustrative of, it is intended to be used for explaining the application, and it is not intended that restriction to the application.
Below with reference to the accompanying drawings completion search method based on artificial intelligence and the device of the embodiment of the present application are described.
Fig. 1 is the flow chart of the completion search method based on artificial intelligence of one embodiment of the application.
As it is shown in figure 1, should include based on the completion search method of artificial intelligence:
Step 101, receives the first query statement that user is currently entered, judges whether described first query statement meets completion condition according to default completion search strategy.
Specifically, the completion search method based on artificial intelligence that the embodiment of the present invention provides is applied to be had in the terminal unit of search function.Wherein, the type of terminal unit is a lot, for instance: smart mobile phone, panel computer, computer etc..
When receiving the first query statement that user is currently entered, according to default completion search strategy, this first query statement is analyzed, it is judged that whether this first query statement meets completion condition.If this first query statement is unsatisfactory for completion condition, then this first query statement is made directly retrieval, and retrieves result to user feedback.If this first query statement meets completion condition, then the second query statement according to a upper input determines whether this first query statement is carried out completion retrieval.
It should be noted that can need to arrange different completion search strategies according to different application, for instance:
Example one: can be analyzed judging whether this first query statement meets completion condition to the first query statement according to default phrase structure；And/or,
Example two: can according to the whether input inquiry statement of user in the preset time period before current time, it is judged that whether this first query statement meets completion condition.
It should be noted that above-mentioned completion search strategy is merely illustrative, it is possible to need be configured or adjust according to practical application.
Step 102, if described first query statement meets completion condition, then judges that whether the second query statement of the upper retrieval of described user is relevant to described first query statement.
Specifically, if the first query statement meets default completion condition, then judge that whether the second query statement of the upper retrieval of user is relevant to this first query statement, if the second query statement is relevant to this first query statement, then the first query statement is carried out completion retrieval；If the second query statement is uncorrelated with this first query statement, then directly the first query statement is retrieved.
It should be noted that can need in different ways according to different application, it is judged that whether the second query statement of the upper retrieval of user is relevant to the first query statement of current retrieval, for instance:
Example one: the second query statement and the degree of association of the first query statement can be assessed according to the discrimination model of training in advance, if the degree of association meets default threshold value, it is determined that the second query statement and the first query statement are correlated with.
Example two: can judge whether the second query statement and the first query statement are correlated with according to the phrase structure related information prestored.
It should be noted that the above-mentioned judgment mode whether the first query statement and the second query statement are correlated with is merely illustrative, it is possible to need be configured or adjust according to practical application.
Step 103, if described second query statement is relevant to described first query statement, then carries out completion according to described second query statement to described first query statement and generates the 3rd query statement.
Step 104, retrieves described 3rd query statement, and retrieves result to described user feedback.
Specifically, if it is determined that know that the second query statement of the upper retrieval of user is relevant to the first query statement of current retrieval, then according to the second query statement, the first query statement of current retrieval is carried out completion and generate the 3rd query statement.
And then, the 3rd query statement after completion is retrieved, and retrieves result to user feedback.
According to default completion search strategy, the completion search method based on artificial intelligence of the embodiment of the present application, by receiving the first query statement that user is currently entered, judges whether described first query statement meets completion condition；If described first query statement meets completion condition, then judge that whether the second query statement of the upper retrieval of described user is relevant to described first query statement；If described second query statement is relevant to described first query statement, then according to described second query statement, described first query statement is carried out completion and generate the 3rd query statement；Described 3rd query statement is retrieved, and retrieves result to described user feedback.Hereby it is achieved that improved search engines and the mutual efficiency of user by many wheel retrieval modes, save the time of user input query statement.
Fig. 2 is the flow chart of the completion search method based on artificial intelligence of another embodiment of the application.
As in figure 2 it is shown, should may comprise steps of based on the completion search method of artificial intelligence:
Step 201, receives the first query statement that user is currently entered.
Step 202, it is judged that whether described first query statement comprises default phrase structure.
Specifically, pre-set the phrase structure whether meeting completion condition for detecting the first query statement, the compound mode of phrase structure can need to be configured according to practical application, such as: the phrase structure preset includes: entity word and attribute word, when the first query statement is " the New York mayor ", then knowing that this query statement includes the phrase structure preset, wherein, " New York " is entity word；" mayor " is attribute word.When the first query statement is " color ", then know that this query statement does not include the phrase structure preset.
If described first query statement does not comprise default phrase structure, then described first query statement meets completion condition, and then performs step 203；If described first query statement comprises default phrase structure, then described first query statement is unsatisfactory for completion condition, then perform step 207；
Step 203, it is judged that described user whether input inquiry statement in the preset time period before current time.
Specifically, pre-set a time threshold, judge in from current time to the preset time period of this time threshold described user whether input inquiry statement, to determine whether there is the second query statement that the first query statement can carry out completion, and then judge whether the first query statement meets completion condition.
If it is determined that be informed in this preset time period described user input query statement, then described first query statement meets completion condition, and then performs step 204；If it is determined that be informed in described preset time period described user do not have input inquiry statement, then described first query statement is unsatisfactory for completion condition, and then performs step 207；
Step 204, it is judged that whether the second query statement of the upper retrieval of user comprises default phrase structure；
Specifically, if the first query statement meets completion condition, then judge that whether the second query statement of the upper detection of the first query statement and user is relevant.First, it is determined that whether this second query statement comprises default phrase structure.Wherein, the compound mode of the phrase structure pre-set can need to be configured according to practical application.
If the second query statement comprises default phrase structure, then know that this second query statement and the first query statement are uncorrelated, and then perform step 207；If this second query statement does not comprise default phrase structure, and then performs step 205；
According to default phrase relation storehouse, step 205, judges the first phrase in described first query statement is with whether the second phrase in described second query statement has dependency,
Specifically, pre-set phrase relation storehouse, wherein it is possible to need to arrange the dependency of each phrase structure in this phrase relation storehouse according to practical application, such as: this phrase relation storehouse includes entity dictionary and attribute dictionary, and the corresponding relation of each word of entity dictionary and each word in attribute dictionary.
If phrase relation storehouse comprises the first phrase in the first query statement and the second phrase in described second query statement, then illustrate that the second query statement is relevant to described first query statement, and then perform step 206；If phrase relation storehouse does not comprise the first phrase in the first query statement and the second phrase in described second query statement, then illustrate that the second query statement and described first query statement are uncorrelated, and then perform step 207.
Step 206, carries out completion according to described second query statement to described first query statement and generates the 3rd query statement, described 3rd query statement is retrieved, and retrieve result to described user feedback.
Specifically, if it is determined that know that the second query statement of the upper retrieval of user is relevant to the first query statement of current retrieval, then according to the second query statement, the first query statement of current retrieval is carried out completion and generate the 3rd query statement.
And then, the 3rd query statement after completion is retrieved, and retrieves result to user feedback.
Step 207, retrieves described first query statement, and retrieves result to described user feedback.
Specifically, if the first query statement is unsatisfactory for completion condition, or, it is unsatisfactory for dependency between the first query statement and the second query statement, then directly this first query statement is retrieved, and retrieve result to described user feedback.
According to default completion search strategy, the completion search method based on artificial intelligence of the embodiment of the present application, by receiving the first query statement that user is currently entered, judges whether described first query statement meets completion condition；If described first query statement meets completion condition, then judge that whether the second query statement of the upper retrieval of described user is relevant to described first query statement；If described second query statement is relevant to described first query statement, then according to described second query statement, described first query statement is carried out completion and generate the 3rd query statement；Described 3rd query statement is retrieved, and retrieves result to described user feedback.Hereby it is achieved that improved search engines and the mutual efficiency of user by many wheel retrieval modes, save the time of user input query statement.
For the implementation process of the completion search method based on artificial intelligence that explanation the application clearly provides, it is described in detail by following example.
Fig. 3 is based on entity attribute knowledge base and takes turns the system-wide technology frame chart of benefit more.As it is shown on figure 3, have a critically important problem is how to judge that the query statement that user inputs needs to carry out completion retrieval in based on the completion search method of artificial intelligence.First, judge whether described first query statement meets completion condition according to default completion search strategy；If described first query statement meets completion condition, then judge that whether the second query statement of the upper retrieval of described user is relevant to described first query statement.
How to judge that whether the query statement that user inputs is relevant with previous query statement, namely judge whether to need to use the current queries statement that user is inputted by previous query statement to carry out completion.Such as previous query statement is " San Francisco mayor ", the current queries statement of user's input then needs completion to become " San Francisco area " for " area ", but the current queries statement of user's input is " Liu Dehua ", and then query statement without completion or after completion is itself, because user opens a new search scene " Liu Dehua ", this query statement also just becomes first query statement of other query statement follow-up.
Judging whether to need the policy mandates of completion to have significantly high precision, because completion mistake can cause the Consumer's Experience of very severe, proposing of novelty uses the method for entity attribute knowledge base to judge whether to need completion herein.Entity attribute knowledge base is all entity word and the relational knowledge base of attribute word, may determine that whether given entity word (such as " San Francisco ") and given attribute word (such as " mayor ") have relation according to this knowledge base, therefore need during completion first to identify a upper query statement of user's input and the entity word of current statement and attribute word respectively, recognition methods uses the Forward Maximum Method method in existing participle technique, if a upper query statement or current queries statement also have other word except entity word and attribute word, then not within the method solution scope, namely current queries statement is not carried out completion.After identifying entity word and attribute word, if namely current statement has entity word and attribute word, then it is assumed that current sentence composition is complete in completion.Current statement only has entity word A and when previous statement has attribute word B, it is judged that whether A and B has relation, has, and the attribute word using a statement carrys out the entity word of completion current statement；When current statement only has attribute word B and when previous statement has entity word A, it is judged that whether A and B has relation, has, the entity word using a statement carrys out the attribute word of completion current statement.
Fig. 4 is the schematic diagram being realized many wheel search by completion retrieval mode.Referring to Fig. 4, in the session of many wheels, a critically important problem is that after user inputs a query statement, how the information of previous query statement or even multiple queries statement is joined in current queries by search engine, the information completion being achieved in that the query statement previous query statement of use to user's input herein becomes complete query statement, search engine only use completion after query statement carry out retrieving, it is analyzed without in conjunction with previous or even front several query statements, namely query statement and traditional single-wheel search engine after using completion just can realize taking turns search engine more, the core searching system of search engine and search strategy are without any change.
Require emphasis be a bit that user input query statement is carried out completion time what use is not previous original query statement, but the query statement after previous completion.In Fig. 5, user is inputted the 3rd query statement when carrying out completion, what use is not second original query statement " mayor ", but the query statement " San Francisco mayor " after second completion, if in like manner there being the 4th query statement, that alignment completion is the query statement after using the 3rd completion, the completion of such recursion can constantly be delivered to follow-up all query statements by taking turns scene " San Francisco " from first query statement easily, and the query statement that user is inputted has only to the query statement after using previous completion when carrying out completion, without using front several or first query statement.
Judge whether the IP initiating query statement or user in the first few minutes initiating this query statement (being usually 10 minutes) have the query statement initiating other, then this query statement is not first query statement, and the sentence after first query statement completion is forced to be set as itself.
Fig. 5 is an exemplary plot of traditional single-wheel search engine.Referring to Fig. 5, user inputs a query statement, search engine provides Query Result, user inputs a query statement again, the Query Result made new advances given by search engine, before and after user, two query statements of input are at search engine it appear that incoherent, and namely first Query Result is only correlated with first query statement, and second Query Result is only correlated with second query statement.
Fig. 6 is that the schematic diagram supplementing retrieval taken turns by search engine more.Referring to Fig. 6, when user inputs first query statement, search engine provides Query Result, when user inputs a query statement again, search engine can according to this query statement and in conjunction with previous query statement even before several query statements provide Query Result, the conversational search interactive mode of this many wheels is truly realized the mutual of displaying, schematic diagram interacts round city " San Francisco " scene, first query statement gives interaction scenarios, follow-up query statement is no need for scene is repeated explanation, only need to indicate the inquiry content under this scene, namely when user needs inquiry " San Francisco mayor ", only need input " mayor ", the input time of user 60% can be saved.
In order to realize above-described embodiment, the application also proposes a kind of completion based on artificial intelligence and retrieves device.
Fig. 7 is the structural representation of the completion retrieval device based on artificial intelligence of one embodiment of the application.
Included as it is shown in fig. 7, device should be retrieved based on the completion of artificial intelligence:
Receiver module 11, for receiving the first query statement that user is currently entered；
According to the completion search strategy preset, first judge module 12, for judging whether described first query statement meets completion condition；
Second judge module 13, if meeting completion condition for described first query statement, then judges that whether the second query statement of the upper retrieval of described user is relevant to described first query statement；
Completion module 14, if relevant to described first query statement for described second query statement, then carry out completion according to described second query statement to described first query statement and generates the 3rd query statement；
Retrieval module 15, for described 3rd query statement is retrieved, and retrieves result to described user feedback.
It should be noted that device is retrieved in the completion based on artificial intelligence that the aforementioned explanation to the completion search method embodiment based on artificial intelligence is also applied for this embodiment, repeat no more herein.
Device is retrieved in the completion based on artificial intelligence of the embodiment of the present application, by receiving the first query statement that user is currently entered, judges whether described first query statement meets completion condition according to default completion search strategy；If described first query statement meets completion condition, then judge that whether the second query statement of the upper retrieval of described user is relevant to described first query statement；If described second query statement is relevant to described first query statement, then according to described second query statement, described first query statement is carried out completion and generate the 3rd query statement；Described 3rd query statement is retrieved, and retrieves result to described user feedback.Hereby it is achieved that improved search engines and the mutual efficiency of user by many wheel retrieval modes, save the time of user input query statement.
Based on above-described embodiment, further, described retrieval module 15 is additionally operable to:
If described first query statement is unsatisfactory for completion condition, then described first query statement is retrieved, and retrieve result to described user feedback.
Further, described retrieval module 15 is additionally operable to:
If described second query statement is uncorrelated with described first query statement, then described first query statement is retrieved, and retrieve result to described user feedback.
In one embodiment, described first judge module 12 is used for:
Judge whether described first query statement comprises default phrase structure；
If described first query statement does not comprise default phrase structure, then judge in the preset time period before current time described user whether input inquiry statement；
If it is determined that be informed in described preset time period described user input query statement, then described first query statement meets completion condition.
Wherein, described default phrase structure includes:
Entity word and attribute word.
Further, described first judge module 12 is additionally operable to:
If described first query statement comprises default phrase structure, then described first query statement is unsatisfactory for completion condition.
Further, described first judge module 12 is additionally operable to:
If it is determined that be informed in described preset time period described user do not have input inquiry statement, then described first query statement is unsatisfactory for completion condition.
In one embodiment, described second judge module 13 is used for:
Judge whether described second query statement comprises default phrase structure；
If described second query statement comprises default phrase structure, then described second query statement is uncorrelated with described first query statement.
Further, described second judge module 13 is additionally operable to:
If described second query statement does not comprise default phrase structure, then judge the first phrase in described first query statement is with whether the second phrase in described second query statement has dependency according to the phrase relation storehouse preset；
If described first phrase and described second phrase have dependency, then described second query statement is relevant to described first query statement；
If described first phrase and described second phrase do not have dependency, then described second query statement is uncorrelated with described first query statement.
Wherein, described default phrase relation storehouse, including:
Entity dictionary and attribute dictionary, and the corresponding relation of each word of entity dictionary and each word of attribute dictionary.
Device is retrieved in the completion based on artificial intelligence of the embodiment of the present application, by receiving the first query statement that user is currently entered, judges whether described first query statement meets completion condition according to default completion search strategy；If described first query statement meets completion condition, then judge that whether the second query statement of the upper retrieval of described user is relevant to described first query statement；If described second query statement is relevant to described first query statement, then according to described second query statement, described first query statement is carried out completion and generate the 3rd query statement；Described 3rd query statement is retrieved, and retrieves result to described user feedback.Hereby it is achieved that improved search engines and the mutual efficiency of user by many wheel retrieval modes, save the time of user input query statement.
In the description of this specification, the description of reference term " embodiment ", " some embodiments ", " example ", " concrete example " or " some examples " etc. means to be contained at least one embodiment or the example of the application in conjunction with specific features or the feature of this embodiment or example description.In this manual, the schematic representation of above-mentioned term is necessarily directed to identical embodiment or example.And, the specific features of description or feature can combine in one or more embodiments in office or example in an appropriate manner.Additionally, when not conflicting, the feature of the different embodiments described in this specification or example and different embodiment or example can be carried out combining and combining by those skilled in the art.
Additionally, term " first ", " second " are only for descriptive purposes, and it is not intended that indicate or imply relative importance or the implicit quantity indicating indicated technical characteristic.Thus, define " first ", the feature of " second " can express or implicitly include at least one this feature.In the description of the present application, " multiple " are meant that at least two, for instance two, three etc., unless otherwise expressly limited specifically.
Describe in flow chart or in this any process described otherwise above or method and be construed as, represent and include the module of code of executable instruction of one or more step for realizing custom logic function or process, fragment or part, and the scope of the preferred implementation of the application includes other realization, wherein can not press order that is shown or that discuss, including according to involved function by basic mode simultaneously or in the opposite order, performing function, this should be understood by embodiments herein person of ordinary skill in the field.
Claims (20)
1. the completion search method based on artificial intelligence, it is characterised in that comprise the following steps:
Receive the first query statement that user is currently entered, judge whether described first query statement meets completion condition according to default completion search strategy；
If described first query statement meets completion condition, then judge that whether the second query statement of the upper retrieval of described user is relevant to described first query statement；
If described second query statement is relevant to described first query statement, then according to described second query statement, described first query statement is carried out completion and generate the 3rd query statement；
Described 3rd query statement is retrieved, and retrieves result to described user feedback.
2. the method for claim 1, it is characterised in that also include:
If described first query statement is unsatisfactory for completion condition, then described first query statement is retrieved, and retrieve result to described user feedback.
3. the method for claim 1, it is characterised in that also include:
If described second query statement is uncorrelated with described first query statement, then described first query statement is retrieved, and retrieve result to described user feedback.
4. the method for claim 1, it is characterised in that the completion search strategy that described basis is preset judges whether described first query statement meets completion condition, including:
Judge whether described first query statement comprises default phrase structure；
If described first query statement does not comprise default phrase structure, then judge in the preset time period before current time described user whether input inquiry statement；
If it is determined that be informed in described preset time period described user input query statement, then described first query statement meets completion condition.
5. method as claimed in claim 4, it is characterised in that described default phrase structure, including:
Entity word and attribute word.
6. method as claimed in claim 4, it is characterised in that the completion search strategy that described basis is preset judges whether described first query statement meets completion condition, also includes:
If described first query statement comprises default phrase structure, then described first query statement is unsatisfactory for completion condition.
7. method as claimed in claim 4, it is characterised in that the completion search strategy that described basis is preset judges whether described first query statement meets completion condition, also includes:
If it is determined that be informed in described preset time period described user do not have input inquiry statement, then described first query statement is unsatisfactory for completion condition.
8. the method as described in as arbitrary in claim 1-7, it is characterised in that described judge that whether the second query statement of the upper retrieval of described user is relevant to described first query statement, including:
Judge whether described second query statement comprises default phrase structure；
If described second query statement comprises default phrase structure, then described second query statement is uncorrelated with described first query statement.
9. method as claimed in claim 8, it is characterised in that described judge that whether the second query statement of the upper retrieval of described user is relevant to described first query statement, also includes:
If described second query statement does not comprise default phrase structure, then judge the first phrase in described first query statement is with whether the second phrase in described second query statement has dependency according to the phrase relation storehouse preset；
If described first phrase and described second phrase have dependency, then described second query statement is relevant to described first query statement；
If described first phrase and described second phrase do not have dependency, then described second query statement is uncorrelated with described first query statement.
10. method as claimed in claim 9, it is characterised in that described default phrase relation storehouse, including:
Entity dictionary and attribute dictionary, and the corresponding relation of each word of entity dictionary and each word of attribute dictionary.
11. device is retrieved in the completion based on artificial intelligence, it is characterised in that including:
Receiver module, for receiving the first query statement that user is currently entered；
According to the completion search strategy preset, first judge module, for judging whether described first query statement meets completion condition；
Second judge module, if meeting completion condition for described first query statement, then judges that whether the second query statement of the upper retrieval of described user is relevant to described first query statement；
Completion module, if relevant to described first query statement for described second query statement, then carry out completion according to described second query statement to described first query statement and generates the 3rd query statement；
Retrieval module, for described 3rd query statement is retrieved, and retrieves result to described user feedback.
12. device as claimed in claim 11, it is characterised in that described retrieval module is additionally operable to:
If described first query statement is unsatisfactory for completion condition, then described first query statement is retrieved, and retrieve result to described user feedback.
13. device as claimed in claim 11, it is characterised in that described retrieval module is additionally operable to:
If described second query statement is uncorrelated with described first query statement, then described first query statement is retrieved, and retrieve result to described user feedback.
14. device as claimed in claim 11, it is characterised in that described first judge module is used for:
Judge whether described first query statement comprises default phrase structure；
If described first query statement does not comprise default phrase structure, then judge in the preset time period before current time described user whether input inquiry statement；
If it is determined that be informed in described preset time period described user input query statement, then described first query statement meets completion condition.
15. device as claimed in claim 14, it is characterised in that described default phrase structure includes:
Entity word and attribute word.
16. device as claimed in claim 14, it is characterised in that described first judge module is additionally operable to:
If described first query statement comprises default phrase structure, then described first query statement is unsatisfactory for completion condition.
17. device as claimed in claim 14, it is characterised in that described first judge module is additionally operable to:
If it is determined that be informed in described preset time period described user do not have input inquiry statement, then described first query statement is unsatisfactory for completion condition.
18. the device as described in as arbitrary in claim 11-17, it is characterised in that described second judge module is used for:
Judge whether described second query statement comprises default phrase structure；
If described second query statement comprises default phrase structure, then described second query statement is uncorrelated with described first query statement.
19. device as claimed in claim 18, it is characterised in that described second judge module is additionally operable to:
If described second query statement does not comprise default phrase structure, then judge the first phrase in described first query statement is with whether the second phrase in described second query statement has dependency according to the phrase relation storehouse preset；
If described first phrase and described second phrase have dependency, then described second query statement is relevant to described first query statement；
If described first phrase and described second phrase do not have dependency, then described second query statement is uncorrelated with described first query statement.
20. device as claimed in claim 19, it is characterised in that described default phrase relation storehouse, including:
Entity dictionary and attribute dictionary, and the corresponding relation of each word of entity dictionary and each word of attribute dictionary.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610119571.6A CN105808688B (en) | 2016-03-02 | 2016-03-02 | Complementary retrieval method and device based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610119571.6A CN105808688B (en) | 2016-03-02 | 2016-03-02 | Complementary retrieval method and device based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105808688A true CN105808688A (en) | 2016-07-27 |
| CN105808688B CN105808688B (en) | 2021-02-05 |
Family
ID=56466590
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610119571.6A Active CN105808688B (en) | 2016-03-02 | 2016-03-02 | Complementary retrieval method and device based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105808688B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106991152A (en) * | 2017-03-28 | 2017-07-28 | 百度在线网络技术（北京）有限公司 | Search method, device and the terminal device of on-demand assets based on artificial intelligence |
| CN108563637A (en) * | 2018-04-13 | 2018-09-21 | 北京理工大学 | A kind of sentence entity complementing method of fusion triple knowledge base |
| CN109857845A (en) * | 2019-01-03 | 2019-06-07 | 北京奇艺世纪科技有限公司 | Model training and data retrieval method, device, terminal and computer readable storage medium |
| CN114090722A (en) * | 2022-01-19 | 2022-02-25 | 支付宝(杭州)信息技术有限公司 | Method and device for automatically completing query content |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005004029A2 (en) * | 2003-07-07 | 2005-01-13 | Woncheol Jeong | Interactive search method. |
| US20070276820A1 (en) * | 2006-05-25 | 2007-11-29 | International Business Machines Corporation | System, method and program for key work searching |
| CN102902753A (en) * | 2012-09-20 | 2013-01-30 | 北京奇虎科技有限公司 | Method and device for complementing search terms and establishing individual interest models |
| CN103383701A (en) * | 2013-07-12 | 2013-11-06 | 北京小米科技有限责任公司 | Information retrieving method, device and terminal |
| CN103678358A (en) * | 2012-09-13 | 2014-03-26 | 腾讯科技（深圳）有限公司 | Information search method and system |
| CN104462505A (en) * | 2014-12-19 | 2015-03-25 | 北京奇虎科技有限公司 | Search method and device |
-
2016
- 2016-03-02 CN CN201610119571.6A patent/CN105808688B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005004029A2 (en) * | 2003-07-07 | 2005-01-13 | Woncheol Jeong | Interactive search method. |
| US20070276820A1 (en) * | 2006-05-25 | 2007-11-29 | International Business Machines Corporation | System, method and program for key work searching |
| CN103678358A (en) * | 2012-09-13 | 2014-03-26 | 腾讯科技（深圳）有限公司 | Information search method and system |
| CN102902753A (en) * | 2012-09-20 | 2013-01-30 | 北京奇虎科技有限公司 | Method and device for complementing search terms and establishing individual interest models |
| CN103383701A (en) * | 2013-07-12 | 2013-11-06 | 北京小米科技有限责任公司 | Information retrieving method, device and terminal |
| CN104462505A (en) * | 2014-12-19 | 2015-03-25 | 北京奇虎科技有限公司 | Search method and device |
Non-Patent Citations (2)
| Title |
|---|
| M. BAROUNI-EBRAHIMI等: "On Query Completion in Web Search Engines Based on Query Stream Mining", 《IEEE》 * |
| 易博: "基于主动学习的语义确实问句补全", 《中国优秀硕士学位论文全文数据库信息科技辑（月刊 ）》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106991152A (en) * | 2017-03-28 | 2017-07-28 | 百度在线网络技术（北京）有限公司 | Search method, device and the terminal device of on-demand assets based on artificial intelligence |
| CN108563637A (en) * | 2018-04-13 | 2018-09-21 | 北京理工大学 | A kind of sentence entity complementing method of fusion triple knowledge base |
| CN109857845A (en) * | 2019-01-03 | 2019-06-07 | 北京奇艺世纪科技有限公司 | Model training and data retrieval method, device, terminal and computer readable storage medium |
| CN114090722A (en) * | 2022-01-19 | 2022-02-25 | 支付宝(杭州)信息技术有限公司 | Method and device for automatically completing query content |
| CN114090722B (en) * | 2022-01-19 | 2022-04-22 | 支付宝(杭州)信息技术有限公司 | Method and device for automatically completing query content |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105808688B (en) | 2021-02-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107291783B (en) | Semantic matching method and intelligent equipment | |
| CN108984650B (en) | Computer-readable recording medium and computer device | |
| CN107818781B (en) | Intelligent interaction method, equipment and storage medium | |
| CN106601237B (en) | Interactive voice response system and voice recognition method thereof | |
| CN111625635A (en) | Question-answer processing method, language model training method, device, equipment and storage medium | |
| CN106649825B (en) | Voice interaction system and creation method and device thereof | |
| CN109710935B (en) | Museum navigation and knowledge recommendation method based on cultural relic knowledge graph | |
| CN109410927A (en) | Offline order word parses the audio recognition method combined, device and system with cloud | |
| CN103106282B (en) | A kind of method of Webpage search and displaying | |
| CN111291210B (en) | Image material library generation method, image material recommendation method and related devices | |
| CN109657054A (en) | Abstraction generating method, device, server and storage medium | |
| CN106202476A (en) | A kind of interactive method and device of knowledge based collection of illustrative plates | |
| JP2020521210A (en) | Information processing method and terminal, computer storage medium | |
| CN109801119B (en) | Interface display method, information providing method, user behavior content information processing method and equipment | |
| CN105808688A (en) | Complementation retrieval method and device based on artificial intelligence | |
| US20220253710A1 (en) | Human-Machine Multi-Turn Conversation Method and System for Human-Machine Interaction, and Intelligent Apparatus | |
| CN104462064A (en) | Method and system for prompting content input in information communication of mobile terminals | |
| CN111708869A (en) | Man-machine conversation processing method and device | |
| CN114691831A (en) | Task-type intelligent automobile fault question-answering system based on knowledge graph | |
| US20170115853A1 (en) | Determining Image Captions | |
| CN109308315A (en) | A kind of collaborative recommendation method based on specialist field similarity and incidence relation | |
| WO2019209501A1 (en) | Session message processing | |
| CN108595406B (en) | User state reminding method and device, electronic equipment and storage medium | |
| CN108345612A (en) | A kind of question processing method and device, a kind of device for issue handling | |
| CN112507139A (en) | Knowledge graph-based question-answering method, system, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |