CN109522479A - Search processing method and device - Google Patents
Search processing method and device Download PDFInfo
- Publication number
- CN109522479A CN109522479A CN201811332817.3A CN201811332817A CN109522479A CN 109522479 A CN109522479 A CN 109522479A CN 201811332817 A CN201811332817 A CN 201811332817A CN 109522479 A CN109522479 A CN 109522479A
- Authority
- CN
- China
- Prior art keywords
- term vector
- search
- vector representation
- word segmentation
- euclidean distance
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (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 discloses a kind of search processing method and device, which includes: the search problem for obtaining user's input;According to search problem, the corresponding first term vector representation of search problem is obtained;According to the corresponding second term vector representation of multiple problems of the first term vector representation and pre-stored multiple question and answer centerings, determined in multiple problems and the matched target problem of search problem.Search processing method and device of the invention, solves the problems, such as in the prior art the technical issues of to inaccuracy, manual sorting heavy workload is matched caused by search progress Database field matching, reach accurate to the matching of search problem, the small technical effect of manual sorting workload.
Description
Technical field
The present invention relates to search field more particularly to a kind of search processing methods and device.
Background technique
Search processing method carries out usually after the search problem for getting user's input with the information prestored in database
Matching, the search result that then will match to return to user.
In the related technology, the search problem usually by user's input carries out Database field matching, then to each field
Weighting will take the search result being matched to after weighting to return to user.
But present inventor discovery at least the foregoing technology has the following technical problems:
Technical problem one: carrying out Database field matching to search problem, can only recognize the field letter in search problem
Breath, can not accurately identify the information of the whole sentence of search problem, cause matching result inaccurate.
Technical problem two: a large amount of field information, such as category, Questions types, key to the issue need to be prepared in the database
Word etc. leads to the heavy workload of manual sorting.
Summary of the invention
The embodiment of the present application solves the problems, such as in the prior art by providing a kind of search processing method and device to search
Using inaccuracy is matched caused by Database field matching way, the problem of manual sorting heavy workload, realize to search
The accurate match of problem reduces the workload of manual sorting.
The application first aspect embodiment provides a kind of search processing method, comprising the following steps:
Obtain the search problem of user's input;
According to described search problem, the corresponding first term vector representation of described search problem is obtained;
According to multiple problems of the first term vector representation and pre-stored multiple question and answer centerings corresponding
Two term vector representations are determined and the matched target problem of described search problem in the multiple problem.
It is described according to the first term vector representation and pre-stored multiple according to one embodiment of the application
The corresponding second term vector representation of multiple problems of question and answer centering, determines to ask with described search in the multiple problem
Inscribe matched target problem, comprising: calculate the first term vector representation and multiple second term vector representations
Between Euclidean distance;According to multiple Euclidean distances, the target problem is determined in the multiple problem.
It is described according to multiple Euclidean distances according to one embodiment of the application, it is determined in the multiple problem
The target problem out, comprising: using problem corresponding to the Euclidean distance the smallest in the multiple problem as the target
Problem.
It is described according to multiple Euclidean distances according to one embodiment of the application, it is determined in the multiple problem
The target problem out, comprising: using problem corresponding to the Euclidean distance the smallest in the multiple problem as candidate problem;
If the smallest Euclidean distance is equal to or less than given threshold, the candidate problem is determined as the target and is asked
Topic.
It is described according to described search problem according to one embodiment of the application, obtain described search problem corresponding the
One term vector representation, comprising: word segmentation processing is carried out to described search problem, obtains multiple first participle results;Obtain institute
State corresponding first term vector of first participle result;By corresponding multiple first term vectors of multiple first participle results
It is added, obtains the first term vector representation.
According to one embodiment of the application, the search processing method further include: divide respectively the multiple problem
Word processing, obtains multiple second word segmentation results;Obtain corresponding second term vector of second word segmentation result;By multiple described
Corresponding multiple second term vectors of two word segmentation results are added, and obtain the second term vector representation.The application second
Aspect embodiment provides a kind of search process device, comprising:
First obtains module, for obtaining the search problem of user's input;
Second obtains module, for obtaining the corresponding first term vector table of described search problem according to described search problem
Show form;
Determining module, for according to the multiple of the first term vector representation and pre-stored multiple question and answer centerings
The corresponding second term vector representation of problem is determined to ask with the matched target of described search problem in the multiple problem
Topic.
According to one embodiment of the application, the determining module includes: computing unit, for calculate first word to
Measure the Euclidean distance between representation and multiple second term vector representations;Determination unit, for according to multiple institutes
Euclidean distance is stated, the target problem is determined in the multiple problem.
According to one embodiment of the application, the determination unit is specifically used for: by institute the smallest in the multiple problem
Problem corresponding to Euclidean distance is stated as the target problem.
According to one embodiment of the application, the determination unit is specifically used for: by institute the smallest in the multiple problem
Problem corresponding to Euclidean distance is stated as candidate problem;If the smallest Euclidean distance is equal to or less than setting threshold
The candidate problem is then determined as the target problem by value.
According to one embodiment of the application, the second acquisition module is specifically used for: dividing described search problem
Word processing, obtains multiple first participle results;Obtain corresponding first term vector of the first participle result;By multiple described
Corresponding multiple first term vectors of one word segmentation result are added, and obtain the first term vector representation.
According to one embodiment of the application, the search process device further include: processing module, for asking the multiple
Topic carries out word segmentation processing respectively, obtains multiple second word segmentation results;Obtain corresponding second term vector of second word segmentation result;
Corresponding multiple second term vectors of multiple second word segmentation results are added, obtaining second term vector indicates shape
Formula.
The application third aspect embodiment provides a kind of electronic equipment, comprising: memory, processor and is stored in described
On memory and the computer program that can run on the processor, when the processor executes described program, realize such as this
Apply for search processing method described in first aspect embodiment.
The application fourth aspect embodiment provides a kind of non-transitorycomputer readable storage medium, is stored thereon with meter
Calculation machine program, which is characterized in that when the program is executed by processor, realize the search as described in the application first aspect embodiment
Processing method.
One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:
1, due to realizing and being searched to search problem using the matched mode of term vector representation is carried out to search problem
Rope, so, the problem for efficiently solving the problems, such as to carry out search matching result inaccuracy caused by Database field matching, into
And realize the accurate match to search problem.
Due to realizing the search to search problem using the matched mode of term vector representation is carried out to search problem,
So efficiently solve need to prepare a large amount of field information in the database caused by the heavy workload of manual sorting ask
Topic, and then realize the advantages of reducing manual sorting workload.
2, by calculating the Euclidean distance between the first term vector representation and multiple second term vector representations, so
The matched target problem of problem is determined and searched for according to Euclidean distance afterwards, it can be achieved that search problem and accurate of target problem
Match.
3, it using problem corresponding to minimum euclidean distance as target problem, can accurately obtain and the matched target of search problem
Problem.
4, the method by minimum euclidean distance compared with given threshold determines target problem, can accurately obtain and search for
The matched target problem of problem.
5, the first participle is obtained after carrying out word segmentation processing to search problem as a result, then acquisition first participle result is corresponding
First term vector, then in such a way that multiple first term vectors that first participle result is corresponding are added, obtain the first word to
Representation is measured, term vector representation corresponding to search problem can accurately, be easily obtained.
6, word segmentation processing is carried out to multiple problems of pre-stored multiple question and answer centerings respectively and obtains multiple second participles
As a result, corresponding second term vector of the second word segmentation result is then obtained, then by the way that multiple second word segmentation results are corresponding more
The mode that a second term vector is added, obtains the second term vector representation, can accurately, easily obtain pre-stored multiple
The corresponding term vector representation of multiple problems of question and answer centering.
Detailed description of the invention
Fig. 1 is the flow chart of search processing method according to an embodiment of the invention;
Fig. 2 is the further refined flow chart to S103 step in embodiment illustrated in fig. 1;
Fig. 3 is the structure chart of search process device according to an embodiment of the invention;
Fig. 4 is the structure chart of electronic equipment according to an embodiment of the invention.
Specific embodiment
The present invention is inaccurate to matching caused by search progress Database field matching in the prior art in order to solve the problems, such as
Really, the problem of manual sorting heavy workload, a kind of search processing method and device are proposed, using to search problem carry out word to
The matching way of representation is measured to realize the search to search problem, realizes the accurate match to search problem, also, subtract
The small workload of manual sorting.
In order to better understand the above technical scheme, the exemplary reality of the disclosure is more fully described below with reference to accompanying drawings
Apply example.Although showing the exemplary embodiment of the disclosure in attached drawing, it being understood, however, that may be realized in various forms this public affairs
It opens and should not be limited by the embodiments set forth herein.It is to be able to thoroughly understand this on the contrary, providing these embodiments
It is open, and the scope of the present disclosure can be fully disclosed to those skilled in the art.
In order to better understand the above technical scheme, in conjunction with appended figures and specific embodiments to upper
Technical solution is stated to be described in detail.
Embodiment one:
Fig. 1 is the flow chart of search processing method according to an embodiment of the invention, as shown in Figure 1, the search process
Method the following steps are included:
S101 obtains the search problem of user's input.
In the embodiment of the present invention, the search problem of user's input is obtained, can be that directly acquire user defeated with written form
The search problem entered is also possible to convert the search problem that user is inputted with speech form, is converted to written form
Search problem.
S102 obtains the corresponding first term vector representation of search problem according to search problem.
In the embodiment of the present invention, according to the search problem of the step S101 user got, it is corresponding to obtain search problem
First term vector representation.
As a kind of feasible embodiment, step 102 be may particularly include: being carried out word segmentation processing to search problem, is obtained
Multiple first participle results;Obtain corresponding first term vector of first participle result;Multiple first participle results are corresponding more
A first term vector is added, and obtains the first term vector representation.
Specifically, doing word segmentation processing to the whole sentence of search problem, such as can be respectively using three kinds points of jieba participle tool
Word mode, i.e. accurate model, syntype, search engine mode carry out word segmentation processing to the whole sentence of search problem, obtain search and ask
Inscribe the word list of multiple participles of the whole sentence under three kinds of participle modes, i.e. first participle result;It is respectively that the participle in list is complete
The multiple first participle results obtained after are input in trained term vector model, and it is corresponding to obtain multiple first participle results
Multiple first term vectors, as a kind of feasible embodiment, term vector model can have been instructed using in TensorFlow
The Chinese word vector model perfected;Corresponding multiple first term vectors of obtained multiple first participle results are added, obtain the
One term vector representation.
S103, according to multiple problems of the first term vector representation and pre-stored multiple question and answer centerings corresponding
Two term vector representations are determined and the matched target problem of search problem in multiple problems.
In the embodiment of the present invention, corresponding second word of multiple problems of multiple question and answer centerings can be stored in advance in the database
Vector representation, the acquisition modes of the second term vector representation can be indicated with the first term vector in S102 step in database
The acquisition modes of form are identical, comprising: carry out word segmentation processing respectively to multiple problems, obtain multiple second word segmentation results;It obtains
Corresponding second term vector of second word segmentation result;Corresponding multiple second term vectors of multiple second word segmentation results are added, are obtained
Second term vector representation, details are not described herein again for detailed process.First term vector representation is asked with multiple in database
It inscribes corresponding second term vector representation to be compared, be indicated with the second term vector that the first term vector representation matches
Problem corresponding to form i.e. and search the matched target problem of problem.
Technical solution in above-mentioned the embodiment of the present application, at least have the following technical effects or advantages:
1, due to realizing and being searched to search problem using the matched mode of term vector representation is carried out to search problem
Rope, so, the problem for efficiently solving the problems, such as to carry out search matching result inaccuracy caused by Database field matching, into
And realize the accurate match to search problem.
Due to realizing the search to search problem using the matched mode of term vector representation is carried out to search problem,
So efficiently solve need to prepare a large amount of field information in the database caused by the heavy workload of manual sorting ask
Topic, and then realize the advantages of reducing manual sorting workload.
2, the first participle is obtained after carrying out word segmentation processing to search problem as a result, then acquisition first participle result is corresponding
First term vector, then in such a way that multiple first term vectors that first participle result is corresponding are added, obtain the first word to
Representation is measured, term vector representation corresponding to search problem can accurately, be easily obtained.
3, word segmentation processing is carried out to multiple problems of pre-stored multiple question and answer centerings respectively and obtains multiple second participles
As a result, corresponding second term vector of the second word segmentation result is then obtained, then by the way that multiple second word segmentation results are corresponding more
The mode that a second term vector is added, obtains the second term vector representation, can accurately, easily obtain pre-stored multiple
The corresponding term vector representation of multiple problems of question and answer centering.
Embodiment two:
Fig. 2 is the further refined flow chart to S103 step in embodiment illustrated in fig. 1, as shown in Fig. 2, shown in Fig. 1
S103 step in embodiment can include:
S201 calculates the Euclidean distance between the first term vector representation and multiple second term vector representations.
In the embodiment of the present invention, multiple second term vector representations in the first term vector representation and database are calculated
Between multiple Euclidean distances.As a kind of feasible embodiment, the numpy in Python is can be used in the calculating of Euclidean distance
Library, the computing rule of multiple Euclidean distances between the first term vector representation and multiple second term vector representations are as follows:
Two n-dimensional vector a (x11,x12,…,x1n) and b (x21,x22,…,x2n) between Euclidean distance
S202 determines target problem in multiple problems according to multiple Euclidean distances.
According between step S201 the first term vector representation being calculated and multiple second term vector representations
Multiple Euclidean distances, determine target problem in multiple problems.
It, can be using problem corresponding to Euclidean distance the smallest in multiple problems as target as a kind of feasible embodiment
Problem.
Specifically, the first term vector representation that step S201 is calculated and multiple second term vector representations
Between multiple Euclidean distances be compared, obtain the smallest Euclidean distance in multiple Euclidean distances, minimum euclidean distance is corresponding
The second term vector representation corresponding to problem i.e. and search the matched target problem of problem.
As another feasible embodiment, using problem corresponding to Euclidean distance the smallest in multiple problems as candidate
Problem;If the smallest Euclidean distance is equal to or less than given threshold, candidate problem is determined as target problem.
Specifically, given threshold can be preset.The first term vector representation that step S201 is calculated and more
Multiple Euclidean distances between a second term vector representation are compared, obtain in multiple Euclidean distances it is the smallest it is European away from
From using problem corresponding to the corresponding second term vector representation of minimum euclidean distance as candidate problem, by the smallest Europe
Formula distance is compared with given threshold, if the smallest Euclidean distance is equal to or less than given threshold, candidate problem is
With the search matched target problem of problem.
Technical solution in above-mentioned the embodiment of the present application, at least have the following technical effects or advantages:
1, by calculating the Euclidean distance between the first term vector representation and multiple second term vector representations, so
The matched target problem of problem is determined and searched for according to Euclidean distance afterwards, it can be achieved that search problem and accurate of target problem
Match.
2, it using problem corresponding to minimum euclidean distance as target problem, can accurately obtain and the matched target of search problem
Problem.
3, the method by minimum euclidean distance compared with given threshold determines target problem, can accurately obtain and search for
The matched target problem of problem.
Based on the same inventive concept, the embodiment of the invention also provides the corresponding device of method in embodiment one and two, see
Embodiment three.
Embodiment three:
Fig. 3 is the structure chart of search process device according to an embodiment of the invention.As shown in figure 3, the search process
Device includes:
First obtains module 21, for obtaining the search problem of user's input;
Second obtains module 22, for obtaining the corresponding first term vector representation of search problem according to search problem;
Determining module 23, for being asked according to the first term vector representation and pre-stored multiple the multiple of question and answer centering
Corresponding second term vector representation is inscribed, is determined in multiple problems and the matched target problem of search problem.
Further, in a kind of possible implementation of the embodiment of the present invention, determining module 23 includes: computing unit,
For calculating the Euclidean distance between the first term vector representation and multiple second term vector representations;Determination unit is used
According to multiple Euclidean distances, target problem is determined in multiple problems.
Further, in a kind of possible implementation of the embodiment of the present invention, determination unit is specifically used for: asking multiple
Problem corresponding to the smallest Euclidean distance is as target problem in topic.
Further, in a kind of possible implementation of the embodiment of the present invention, determination unit is specifically used for: asking multiple
The problem that the smallest Euclidean distance is corresponding in topic is as candidate problem;If the smallest Euclidean distance is equal to or less than setting threshold
Value, then be determined as target problem for candidate problem.
Further, in a kind of possible implementation of the embodiment of the present invention, the second acquisition module 22 is specifically used for: right
Search problem carries out word segmentation processing, obtains multiple first participle results;Obtain corresponding first term vector of first participle result;It will
Corresponding multiple first term vectors of multiple word segmentation results are added, and obtain the first term vector representation.
Further, in a kind of possible implementation of the embodiment of the present invention, the search process device further include: processing
Module obtains multiple second word segmentation results for carrying out word segmentation processing respectively to multiple problems;It is corresponding to obtain the second word segmentation result
The second term vector;Corresponding multiple second term vectors of multiple second word segmentation results are added, obtaining the second term vector indicates shape
Formula.
By the device that the embodiment of the present invention three is introduced, used by the method to implement the embodiment of the present invention one and two
Device, so based on the method that the embodiment of the present invention one and two is introduced, the affiliated personnel in this field can understand the tool of the device
Body structure and deformation, so details are not described herein.Device used by the method for all embodiment of the present invention one and two belongs to
The range of the invention to be protected.
Technical solution in above-mentioned the embodiment of the present application, at least have the following technical effects or advantages:
1, due to realizing and being searched to search problem using the matched mode of term vector representation is carried out to search problem
Rope, so, the problem for efficiently solving the problems, such as to carry out search matching result inaccuracy caused by Database field matching, into
And realize the accurate match to search problem.
Due to realizing the search to search problem using the matched mode of term vector representation is carried out to search problem,
So efficiently solve need to prepare a large amount of field information in the database caused by the heavy workload of manual sorting ask
Topic, and then realize the advantages of reducing manual sorting workload.
2, by calculating the Euclidean distance between the first term vector representation and multiple second term vector representations, so
The matched target problem of problem is determined and searched for according to Euclidean distance afterwards, it can be achieved that search problem and accurate of target problem
Match.
3, it using problem corresponding to minimum euclidean distance as target problem, can accurately obtain and the matched target of search problem
Problem.
4, the method by minimum euclidean distance compared with given threshold determines target problem, can accurately obtain and search for
The matched target problem of problem.
5, the first participle is obtained after carrying out word segmentation processing to search problem as a result, then acquisition first participle result is corresponding
First term vector, then in such a way that multiple first term vectors that first participle result is corresponding are added, obtain the first word to
Representation is measured, term vector representation corresponding to search problem can accurately, be easily obtained.
6, word segmentation processing is carried out to multiple problems of pre-stored multiple question and answer centerings respectively and obtains multiple second participles
As a result, corresponding second term vector of the second word segmentation result is then obtained, then by the way that multiple second word segmentation results are corresponding more
The mode that a second term vector is added, obtains the second term vector representation, can accurately, easily obtain pre-stored multiple
The corresponding term vector representation of multiple problems of question and answer centering.
Based on the same inventive concept, the embodiment of the invention also provides the corresponding electronics of method in embodiment one and two to set
It is standby, see
Example IV.
Example IV:
Fig. 4 is the structure chart of electronic equipment according to an embodiment of the invention.As shown in figure 4, the electronic equipment 50, packet
It includes: memory 51, processor 52 and being stored in the computer program that can be run on memory 51 and on a processor, processor 52
When executing program, to realize the search processing method as shown in above-described embodiment.
Based on the same inventive concept, the embodiment of the invention also provides the corresponding non-transitories of method in embodiment one and two
Computer readable storage medium is shown in embodiment five.
Embodiment five:
The non-transitorycomputer readable storage medium, is stored thereon with computer program, which is executed by processor
When, realize the search processing method as shown in above-described embodiment.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
It should be noted that in the claims, any reference symbol between parentheses should not be configured to power
The limitation that benefit requires.Word "comprising" does not exclude the presence of component or step not listed in the claims.Before component
Word "a" or "an" does not exclude the presence of multiple such components.The present invention can be by means of including several different components
It hardware and is realized by means of properly programmed computer.In the unit claims listing several devices, these are filled
Several in setting, which can be, to be embodied by the same item of hardware.The use of word first, second, and third is not
Indicate any sequence.These words can be construed to title.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (14)
1. a kind of search processing method, which comprises the following steps:
Obtain the search problem of user's input;
According to described search problem, the corresponding first term vector representation of described search problem is obtained;
According to corresponding second word of multiple problems of the first term vector representation and pre-stored multiple question and answer centerings
Vector representation is determined and the matched target problem of described search problem in the multiple problem.
2. search processing method according to claim 1, which is characterized in that described to indicate shape according to first term vector
The corresponding second term vector representation of multiple problems of formula and pre-stored multiple question and answer centerings, in the multiple problem
It determines and the matched target problem of described search problem, comprising:
Calculate the Euclidean distance between the first term vector representation and multiple second term vector representations;
According to multiple Euclidean distances, the target problem is determined in the multiple problem.
3. control method according to claim 2, which is characterized in that it is described according to multiple Euclidean distances, described
The target problem is determined in multiple problems, comprising:
Using problem corresponding to the Euclidean distance the smallest in the multiple problem as the target problem.
4. control method according to claim 2, which is characterized in that it is described according to multiple Euclidean distances, described
The target problem is determined in multiple problems, comprising:
Using problem corresponding to the Euclidean distance the smallest in the multiple problem as candidate problem;
If the smallest Euclidean distance is equal to or less than given threshold, the candidate problem is determined as the mesh
Mark problem.
5. control method according to claim 1, which is characterized in that it is described according to described search problem, it is searched described in acquisition
The corresponding first term vector representation of Suo Wenti, comprising:
Word segmentation processing is carried out to described search problem, obtains multiple first participle results;
Obtain corresponding first term vector of the first participle result;
Corresponding multiple first term vectors of multiple first participle results are added, obtaining first term vector indicates
Form.
6. search processing method according to claim 1, which is characterized in that further include:
Word segmentation processing is carried out to the multiple problem respectively, obtains multiple second word segmentation results;
Obtain corresponding second term vector of second word segmentation result;
Corresponding multiple second term vectors of multiple second word segmentation results are added, obtaining second term vector indicates
Form.
7. a kind of search process device characterized by comprising
First obtains module, for obtaining the search problem of user's input;
Second obtains module, for according to described search problem, obtaining corresponding first term vector of described search problem to indicate shape
Formula;
Determining module, for multiple problems according to the first term vector representation and pre-stored multiple question and answer centerings
Corresponding second term vector representation is determined and the matched target problem of described search problem in the multiple problem.
8. search process device according to claim 7, which is characterized in that the determining module includes:
Computing unit, for calculating between the first term vector representation and multiple second term vector representations
Euclidean distance;
Determination unit, for determining the target problem in the multiple problem according to multiple Euclidean distances.
9. search process device according to claim 8, which is characterized in that the determination unit is specifically used for:
Using problem corresponding to the Euclidean distance the smallest in the multiple problem as the target problem.
10. search process device according to claim 8, which is characterized in that the determination unit is specifically used for:
Using problem corresponding to the Euclidean distance the smallest in the multiple problem as candidate problem;
If the smallest Euclidean distance is equal to or less than given threshold, the candidate problem is determined as the mesh
Mark problem.
11. search process device according to claim 7, which is characterized in that the second acquisition module is specifically used for:
Word segmentation processing is carried out to described search problem, obtains multiple first participle results;
Obtain corresponding first term vector of the first participle result;
Corresponding multiple first term vectors of multiple first participle results are added, obtaining first term vector indicates
Form.
12. search process device according to claim 7, which is characterized in that further include:
Processing module obtains multiple second word segmentation results for carrying out word segmentation processing respectively to the multiple problem;Described in acquisition
Corresponding second term vector of second word segmentation result;By the corresponding multiple second term vector phases of multiple second word segmentation results
Add, obtains the second term vector representation.
13. a kind of electronic equipment characterized by comprising memory, processor and be stored on the memory and can be in institute
The computer program run on processor is stated, when the processor executes described program, is realized such as any one of claim 1-6 institute
The search processing method stated.
14. a kind of non-transitorycomputer readable storage medium, is stored thereon with computer program, which is characterized in that the program
When being executed by processor, search processing method as claimed in any one of claims 1 to 6 is realized.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811332817.3A CN109522479A (en) | 2018-11-09 | 2018-11-09 | Search processing method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811332817.3A CN109522479A (en) | 2018-11-09 | 2018-11-09 | Search processing method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109522479A true CN109522479A (en) | 2019-03-26 |
Family
ID=65773724
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811332817.3A Pending CN109522479A (en) | 2018-11-09 | 2018-11-09 | Search processing method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109522479A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104899188A (en) * | 2015-03-11 | 2015-09-09 | 浙江大学 | Problem similarity calculation method based on subjects and focuses of problems |
CN105302882A (en) * | 2015-10-14 | 2016-02-03 | 东软集团股份有限公司 | Keyword obtaining method and apparatus |
CN106610950A (en) * | 2016-09-29 | 2017-05-03 | 四川用联信息技术有限公司 | Improved text similarity solution method |
CN108536708A (en) * | 2017-03-03 | 2018-09-14 | 腾讯科技(深圳)有限公司 | A kind of automatic question answering processing method and automatically request-answering system |
-
2018
- 2018-11-09 CN CN201811332817.3A patent/CN109522479A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104899188A (en) * | 2015-03-11 | 2015-09-09 | 浙江大学 | Problem similarity calculation method based on subjects and focuses of problems |
CN105302882A (en) * | 2015-10-14 | 2016-02-03 | 东软集团股份有限公司 | Keyword obtaining method and apparatus |
CN106610950A (en) * | 2016-09-29 | 2017-05-03 | 四川用联信息技术有限公司 | Improved text similarity solution method |
CN108536708A (en) * | 2017-03-03 | 2018-09-14 | 腾讯科技(深圳)有限公司 | A kind of automatic question answering processing method and automatically request-answering system |
Non-Patent Citations (1)
Title |
---|
李安: "《语料库语言学及Python实现》", 31 August 2018, 山东大学出版社 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107329949B (en) | Semantic matching method and system | |
CN111382255B (en) | Method, apparatus, device and medium for question-answering processing | |
CN111738016A (en) | Multi-intention recognition method and related equipment | |
CN110263133B (en) | Knowledge graph-based question and answer method, electronic device, equipment and storage medium | |
CN110489424A (en) | A kind of method, apparatus, storage medium and the electronic equipment of tabular information extraction | |
CN109857873A (en) | The method and apparatus of recommended entity, electronic equipment, computer-readable medium | |
CN105786898B (en) | A kind of construction method and device of domain body | |
CN109872026A (en) | Evaluation result generation method, device, equipment and computer readable storage medium | |
CN106919551A (en) | A kind of analysis method of emotion word polarity, device and equipment | |
CN109189892A (en) | A kind of recommended method and device based on article review | |
CN112420125A (en) | Molecular attribute prediction method and device, intelligent equipment and terminal | |
CN113220854B (en) | Intelligent dialogue method and device for machine reading and understanding | |
CN110532562A (en) | Neural network training method, Chinese idiom misuse detection method, device and electronic equipment | |
CN114490926A (en) | Method and device for determining similar problems, storage medium and terminal | |
CN113705792A (en) | Personalized recommendation method, device, equipment and medium based on deep learning model | |
CN112632254A (en) | Conversation state determining method, terminal device and storage medium | |
KR101745874B1 (en) | System and method for a learning course automatic generation | |
Jagadamba | Online subjective answer verifying system using artificial intelligence | |
CN106598935B (en) | A kind of method and device of determining document emotion tendency | |
CN109522479A (en) | Search processing method and device | |
CN108021985A (en) | A kind of model parameter training method and device | |
CN112541557B (en) | Training method and device for generating countermeasure network and electronic equipment | |
CN116595189A (en) | Zero sample relation triplet extraction method and system based on two stages | |
CN113704452B (en) | Data recommendation method, device, equipment and medium based on Bert model | |
CN113420545B (en) | Abstract generation method, device, equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190326 |