CN110059253A - A kind of sort method and system and equipment based on natural language analysis - Google Patents
A kind of sort method and system and equipment based on natural language analysis Download PDFInfo
- Publication number
- CN110059253A CN110059253A CN201910331228.1A CN201910331228A CN110059253A CN 110059253 A CN110059253 A CN 110059253A CN 201910331228 A CN201910331228 A CN 201910331228A CN 110059253 A CN110059253 A CN 110059253A
- Authority
- CN
- China
- Prior art keywords
- keyword
- creation
- characteristic
- extraction
- search
- 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.)
- Withdrawn
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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/31—Indexing; Data structures therefor; Storage structures
- G06F16/316—Indexing structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- 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)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of sort methods based on natural language analysis and system and equipment.Wherein, the described method includes: obtaining the input information being input in search engine in search box, with use natural language analysis mode, keyword is carried out to the input information of the acquisition and/or fuzzy word extracts, with to the extraction keyword and/or fuzzy word be normalized, so that the keyword of the extraction and/or the corresponding data value range of each feature of fuzzy word are consistent, form the keyword of the extraction and/or the characteristic of fuzzy word, with the characteristic creation characteristic index to the formation, and it is indexed according to the characteristic of the creation, search matches the data information of the characteristic index of the creation in index data base, and the search result of the data information of the characteristic index for the matching creation for searching out this is shown by relational degree taxis.The degree of association is shown by the above-mentioned means, can be realized and improve search results ranking.
Description
Technical field
The present invention relates to search technique field more particularly to a kind of sort method and system based on natural language analysis with
And equipment.
Background technique
So-called keyword is exactly the text being input in search box when people look for information using engine.Such as it is " happy
Paddy " is exactly keyword, and in addition " Shenzhen joy paddy ", " happy paddy official website " etc. are all keywords, this class keywords can be called
Compound keyword.
So-called Keywords matching degree refers to the same degree of contained keyword in keyword and content of pages, i.e. search is crucial
Word and the title in article or the degree that is consistent in content, matching degree is higher, is more conducive to keyword ranking.
There are mainly two types of viewpoints for the definition of so-called fuzzy matching:
A kind of viewpoint be system allow searched information and search put question between have a certain difference, this species diversity is exactly
The meaning of " fuzzy " in the search.For example, similar Smithe, Smythe will be found out when searching name Smith,
Smyth, Smitt etc..
Another viewpoint is the synonym search that substantial search system carries out automatically.Synonym by system management field
Face configuration.For example, configuration " computer " and " computer " are search " computer ", then comprising " computer " after synonym
Webpage may also appear in search result.
Nowadays keyword match technique and fuzzy matching technology are searching systems mainly by the way of, and this mode has
It is following insufficient:
Search results ranking shows that the degree of association is little.Search results ranking displaying is judged only by keyword, is arranged
The sequence degree of association is little, causes user that can not need in part fast by showing that sequence is quickly found out the data information of needs
Effect is limited in the fast scene accurately retrieved mass data and show related content.
Summary of the invention
In view of this, it is an object of the invention to propose a kind of sort method based on natural language analysis and system and
Equipment can be realized and improve the search results ranking displaying degree of association.
According to an aspect of the present invention, a kind of sort method based on natural language analysis is provided, comprising:
Obtain the input information being input in search engine in search box;
Using natural language analysis mode, keyword is carried out to the input information of the acquisition and/or fuzzy word extracts;
The keyword and/or fuzzy word of the extraction are normalized so that the keyword of the extraction and/or
The corresponding data value range of each feature of fuzzy word is consistent, forms the keyword of the extraction and/or the feature of fuzzy word
Data;
The characteristic creation characteristic index of keyword and/or fuzzy word to the extraction of the formation;
It is indexed according to the characteristic of the creation, search matches the characteristic rope of the creation in index data base
The data information drawn, and the search result of the data information of the characteristic index for the matching creation that described search goes out is pressed
Relational degree taxis is shown.
Wherein, the keyword and/or fuzzy word to the extraction is normalized, so that the pass of the extraction
The corresponding data value range of each feature of keyword and/or fuzzy word is consistent, forms the keyword of the extraction and/or obscures
The characteristic of word, comprising:
At least one feature is extracted from the keyword of the extraction and/or fuzzy word, at least one of the extraction
Feature is normalized, so that the corresponding data value model of each feature of the keyword of the extraction and/or fuzzy word
It encloses unanimously, forms the keyword of the extraction and/or the characteristic of fuzzy word.
Wherein, described to be indexed according to the characteristic of the creation, search matches the creation in index data base
The data information of characteristic index, and the data information of the characteristic index for the matching creation that described search is gone out
Search result is shown by relational degree taxis, comprising:
It is indexed according to the characteristic of the creation and defines search result structure and rule, and the search according to the definition
Resultative construction and rule, search matches the data information of the characteristic index of the creation in index data base, and by institute
The search result for stating the data information of the characteristic index of the matching creation searched out is shown by relational degree taxis.
Wherein, it is indexed described according to the characteristic of the creation, search matches the creation in index data base
Characteristic index data information, and the data information of the characteristic index for the matching creation that described search is gone out
Search result by relational degree taxis show after, further includes:
To user's push and the associated business information of search result shown by relational degree taxis.
According to another aspect of the present invention, a kind of ordering system based on natural language analysis is provided, which is characterized in that
Include:
Obtain module, abstraction module, normalization module, creation module and display module;
The acquisition module, for obtaining the input information being input in search engine in search box;
The abstraction module carries out keyword to the input information of the acquisition for using natural language analysis mode
And/or fuzzy word extracts;
The normalization module, for the extraction keyword and/or fuzzy word be normalized so that institute
The corresponding data value range of each feature of the keyword and/or fuzzy word of stating extraction is consistent, forms the key of the extraction
The characteristic of word and/or fuzzy word;
The creation module, for the keyword of the extraction to the formation and/or the characteristic wound of fuzzy word
Build characteristic index;
The display module, for being indexed according to the characteristic of the creation, the search matching institute in index data base
State the data information of the characteristic index of creation, and the number of the characteristic index for the matching creation that described search is gone out
It is believed that the search result of breath is shown by relational degree taxis.
Wherein, the normalization module, is specifically used for:
At least one feature is extracted from the keyword of the extraction and/or fuzzy word, at least one of the extraction
Feature is normalized, so that the corresponding data value model of each feature of the keyword of the extraction and/or fuzzy word
It encloses unanimously, forms the keyword of the extraction and/or the characteristic of fuzzy word.
Wherein, the display module, is specifically used for:
It is indexed according to the characteristic of the creation and defines search result structure and rule, and the search according to the definition
Resultative construction and rule, search matches the data information of the characteristic index of the creation in index data base, and by institute
The search result for stating the data information of the characteristic index of the matching creation searched out is shown by relational degree taxis.
Wherein, the ordering system based on natural language analysis, further includes:
Pushing module, for believing to user's push and the associated business of the search result shown by relational degree taxis
Breath.
According to a further aspect of the invention, a kind of sequencing equipment based on natural language analysis is provided, which is characterized in that
Include:
At least one processor;And
The memory being connect at least one described processor communication;Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
A processor executes, so that at least one described processor is able to carry out dividing as described in above-mentioned any one based on natural language
The sort method of analysis.
According to a further aspect of the invention, a kind of computer readable storage medium is provided, computer program is stored with,
It is characterized in that, the computer program realizes the row described in any of the above embodiments based on natural language analysis when being executed by processor
Sequence method.
It can be found that above scheme, the available input information being input in search engine in search box, and using certainly
Right language analysis mode carries out keyword to the input information of the acquisition and/or fuzzy word extracts, and to the keyword of the extraction
And/or fuzzy word is normalized, so that the corresponding data of each feature of the keyword of the extraction and/or fuzzy word take
It is consistent to be worth range, forms the keyword of the extraction and/or the characteristic of fuzzy word, and the keyword of the extraction to the formation
And/or the characteristic creation characteristic index of fuzzy word, and indexed according to the characteristic of the creation, in index data
Search matches the data information of the characteristic index of the creation, and the characteristic for matching the creation that this is searched out in library
The search result of the data information of index is shown by relational degree taxis, be can be realized and is improved the search results ranking displaying degree of association.
Further, above scheme can extract at least one feature from the keyword of extraction and/or fuzzy word, right
At least one feature of the extraction is normalized, so that each feature pair of the keyword of the extraction and/or fuzzy word
The data value range answered is consistent, forms the keyword of the extraction and/or the characteristic of fuzzy word, such benefit is can
The calculation amount of keyword and/or fuzzy word is reduced, the search efficiency searched for based on keyword and/or fuzzy word is improved.
Further, above scheme can index according to the characteristic of creation and define search result structure and rule, and
According to the search result structure and rule of this definition, search matches the number of the characteristic index of the creation in index data base
It is believed that breath, and by the search result of the data information of the characteristic index for matching the creation searched out by relational degree taxis
It shows, such benefit is can to define different search result structure and rule according to the demand of user, so that is shown searches
Hitch fruit meets the different demands of user.
Further, above scheme can push and the associated quotient of search result by relational degree taxis displaying to user
Industry information can be realized the business for being input in search engine the input information association in search box with the user to user's push
The precision of information, the business information is high, needed for meeting user, can be improved the success rate of business promotion.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the flow diagram of one embodiment of sort method the present invention is based on natural language analysis;
Fig. 2 is that the present invention is based on the flow diagrams of another embodiment of the sort method of natural language analysis;
Fig. 3 is the structural schematic diagram of one embodiment of ordering system the present invention is based on natural language analysis;
Fig. 4 is that the present invention is based on the structural schematic diagrams of another embodiment of the ordering system of natural language analysis;
Fig. 5 is the structural schematic diagram of the 3rd embodiment of the ordering system the present invention is based on natural language analysis;
Fig. 6 is the structural schematic diagram of one embodiment of sequencing equipment the present invention is based on natural language analysis.
Specific embodiment
With reference to the accompanying drawings and examples, the present invention is described in further detail.It is emphasized that following implement
Example is merely to illustrate the present invention, but is not defined to the scope of the present invention.Likewise, following embodiment is only portion of the invention
Point embodiment and not all embodiments, institute obtained by those of ordinary skill in the art without making creative efforts
There are other embodiments, shall fall within the protection scope of the present invention.
The present invention provides a kind of sort method based on natural language analysis, can be realized and improves search results ranking displaying
The degree of association.
Referring to Figure 1, Fig. 1 is the flow diagram of one embodiment of sort method the present invention is based on natural language analysis.
It is noted that if having substantially the same as a result, method of the invention is not limited with process sequence shown in FIG. 1.Such as Fig. 1
Shown, this method comprises the following steps:
S101: the input information being input in search engine in search box is obtained.
In the present embodiment, the input information of the acquisition being input in search engine in search box can be figure, text
At least one of word, network address etc., the present invention is not limited.
S102: using natural language analysis mode, carries out keyword to the input information of the acquisition and/or fuzzy word is taken out
It takes.
In the present embodiment, the language that natural language can usually develop with culture naturally.For example, English, Chinese, day
Language is the example of natural language, and Esperanto is then fabricated language, is a kind of language created for certain specific purposes.
In the present embodiment, the language that all mankind use includes the above-mentioned language and artificial to develop naturally with culture
Language can be seen as natural language, with the fabricated language set relative to such as programming language etc. for computer, a kind of this nature
Language usage is found in one word of natural language analysis.
In the present embodiment, natural language analysis mode can be syntax and other knowledge with natural language to determine
Composition input each ingredient function of sentence, so as to establishing a kind of data structure and the scheme to obtain input sentence meaning.
In the present embodiment, in compiler theory, the object of natural language analysis mode can be computer program design
The sentence of language;In pattern-recognition, the object of natural language analysis mode can be picture description language isotype language
Sentence, these not instead of sentences of natural language, the sentence of artificial language.In natural language understanding, natural language point
The object of analysis mode can not be artificial language, can be the sentence of natural language.
In the present embodiment, the type of natural language analysis mode may include:
Template matching type analysis program, simple phrase structural grammar analysis program, Transformational Grammar analysis program, extension transfer
Network analysis program, general parsing program, Semantic grammar parsers and without grammer type analysis program etc., the present invention is not
It is limited.
In addition, natural language analysis mode also will do it fine granularity discriminance analysis in the present embodiment, it is more fine to obtain
Classification results.Specifically, normalizing network using posture in the present embodiment to realize fine granularity discriminance analysis.Posture normalization
Network merges the low-level image features such as conv5, fc6 for extracting by posture normalization with unjustified fc8 advanced features, and
The position 2D and 13 semantic position key points are predicted using DPM, or directly using the object frame having been provided and position mark letter
Posture prototype is ceased, and then obtains more sophisticated category result.
S103: being normalized the keyword and/or fuzzy word of the extraction so that the keyword of the extraction and/
Or the corresponding data value range of each feature of fuzzy word is consistent, forms the keyword of the extraction and/or the feature of fuzzy word
Data.
Wherein, which is normalized, so that the keyword of the extraction
And/or the corresponding data value range of each feature of fuzzy word is consistent, forms the keyword of the extraction and/or the spy of fuzzy word
Data are levied, may include:
At least one feature is extracted from the keyword of the extraction and/or fuzzy word, at least one feature of the extraction
It is normalized, so that the corresponding data value range of each feature of the keyword of the extraction and/or fuzzy word is consistent,
The keyword of the extraction and/or the characteristic of fuzzy word are formed, such benefit is to can reduce keyword and/or fuzzy word
Calculation amount, improve the search efficiency searched for based on keyword and/or fuzzy word.
In the present embodiment, normalization can be a kind of mode of simplified calculating, can be the expression formula that will have dimension, pass through
Transformation is crossed, nondimensional expression formula is turned to, becomes scalar.
In the present embodiment, normalization can be a kind of dimensionless processing means, become the absolute value of physical system numerical value
At certain relative value relationship.Normalization can simplify calculating, reduce the effective means of magnitude.For example, each frequency in filter
After value is normalized with cutoff frequency, frequency is all off the relative value of frequency, without dimension.Impedance is returned with internal resistance of source work
One change after, each impedance all at a kind of value of relative impedance, this dimension of ohm be also possible to without.
S104: the characteristic creation characteristic index of keyword and/or fuzzy word to the extraction of the formation.
In the present embodiment, index can be a kind of structure being ranked up to the value of one or more columns per page in database table,
The specific information in database table can be quickly accessed using index.
In the present embodiment, the purpose of creation characteristic index is to speed up lookup or sequence to recording in table.
In the present embodiment, creation characteristic index can greatly improve the performance of system, comprising:
One, pass through creation characteristic index, it is ensured that the uniqueness of every data line in database table.
Two, by creation characteristic index, the retrieval rate of data can be greatly speeded up, this is also creation characteristic
The most important reason of index.
It three, can be with the connection between accelerometer and table, especially in the ginseng for realizing data by creation characteristic index
It is especially significant to examine integrality aspect.
Four, it can equally be shown by creation characteristic index when carrying out data retrieval using grouping and collating sequence clause
Write the time for reducing and being grouped and sorting in inquiry.
Five, by creation characteristic index, device can be hidden using optimizing, improves system during inquiry
Performance.
S105: indexing according to the characteristic of the creation, and search matches the characteristic of the creation in index data base
The data information of index, and by the search result of the data information of the characteristic index for matching the creation searched out by pass
The sequence of connection degree is shown.
Wherein, this according to the creation characteristic index, in index data base search match the creation characteristic
According to the data information of index, and by this search out matching the creation characteristic index data information search result by
Relational degree taxis is shown, may include:
It is indexed according to the characteristic of the creation and defines search result structure and rule, and the search result according to this definition
Structure and rule, search matches the data information of the characteristic index of the creation in index data base, and this is searched out
Matching the creation characteristic index data information search result by relational degree taxis displaying, such benefit is can
To define different search result structure and rule according to the demand of user, so that the search result shown meets the difference of user
Demand.
Wherein, the characteristic index at this according to the creation, search matches the feature of the creation in index data base
The data information of data directory, and by this search out matching the creation characteristic index data information search result
After being shown by relational degree taxis, can also include:
It pushes and should can be realized by the associated business information of search result of relational degree taxis displaying to user to user
Push and the user are input in search engine the business information of the input information association in search box, the business information it is accurate
Degree is high, needed for meeting user, can be improved the success rate of business promotion.
In the present embodiment, the structure and rule of search result can be defined.Such as a characteristic index, Ke Yiyou
The metamessages such as theme, content, storage time, size of data can define different exhibition methods according to user demand and show not
Same element information, does not invent and is not limited.
In the present embodiment, it is catalogue before a book that index data base, which cans be compared to, can accelerate the search speed of database.
It can be found that in the present embodiment, the available input information being input in search engine in search box, and adopt
With natural language analysis mode, keyword is carried out to the input information of the acquisition and/or fuzzy word extracts, and to the pass of the extraction
Keyword and/or fuzzy word are normalized, so that the corresponding number of each feature of the keyword of the extraction and/or fuzzy word
It is consistent according to value range, form the keyword of the extraction and/or the characteristic of fuzzy word, and the pass of the extraction to the formation
The characteristic of keyword and/or fuzzy word creates characteristic index, and is indexed according to the characteristic of the creation, is indexing
Search matches the data information of the characteristic index of the creation, and the feature for matching the creation that this is searched out in database
The search result of the data information of data directory is shown by relational degree taxis, be can be realized and is improved search results ranking displaying association
Degree.
Further, in the present embodiment, at least one spy can be extracted from the keyword of extraction and/or fuzzy word
Sign, is normalized at least one feature of the extraction, so that each spy of the keyword of the extraction and/or fuzzy word
It is consistent to levy corresponding data value range, forms the keyword of the extraction and/or the characteristic of fuzzy word, such benefit is
It can reduce the calculation amount of keyword and/or fuzzy word, improve the search efficiency searched for based on keyword and/or fuzzy word.
Further, in the present embodiment, it can be indexed according to the characteristic of creation and define search result structure and rule
Then, and according to the search result structure and rule of this definition, search matches the characteristic rope of the creation in index data base
The data information drawn, and by the search result of the data information of the characteristic index for matching the creation searched out by association
Degree sequence shows that such benefit is can to define different search result structure and rule according to the demand of user, so that exhibition
The search result shown meets the different demands of user.
Fig. 2 is referred to, Fig. 2 is that the present invention is based on the signals of the process of another embodiment of the sort method of natural language analysis
Figure.In the present embodiment, method includes the following steps:
S201: the input information being input in search engine in search box is obtained.
Can be as above described in S101, therefore not to repeat here.
S202: using natural language analysis mode, carries out keyword to the input information of the acquisition and/or fuzzy word is taken out
It takes.
Can be as above described in S102, therefore not to repeat here.
S203: being normalized the keyword and/or fuzzy word of the extraction so that the keyword of the extraction and/
Or the corresponding data value range of each feature of fuzzy word is consistent, forms the keyword of the extraction and/or the feature of fuzzy word
Data.
Can be as above described in S103, therefore not to repeat here.
S204: the characteristic creation characteristic index of keyword and/or fuzzy word to the extraction of the formation.
Can be as above described in S104, therefore not to repeat here.
S205: indexing according to the characteristic of the creation, and search matches the characteristic of the creation in index data base
The data information of index, and by the search result of the data information of the characteristic index for matching the creation searched out by pass
The sequence of connection degree is shown.
Can be as above described in S105, therefore not to repeat here.
S206: it is pushed to user and is somebody's turn to do the associated business information of search result shown by relational degree taxis.
It can be found that in the present embodiment, can push to user and be associated with the search result shown by relational degree taxis
Business information, can be realized and be input in search engine input information association in search box with the user to user's push
The precision of business information, the business information is high, needed for meeting user, can be improved the success rate of business promotion.
The present invention also provides a kind of ordering system based on natural language analysis, it can be realized and improve search results ranking exhibition
Show the degree of association.
Fig. 3 is referred to, Fig. 3 is the structural schematic diagram of one embodiment of ordering system the present invention is based on natural language analysis.
It should include obtaining module 31, abstraction module 32, normalization module based on the ordering system 30 of natural language analysis in the present embodiment
33, creation module 34 and display module 35.
The acquisition module 31, for obtaining the input information being input in search engine in search box.
The abstraction module 32, for using natural language analysis mode, to the input information of the acquisition carry out keyword and/
Or fuzzy word extracts.
The normalization module 33, for the extraction keyword and/or fuzzy word be normalized so that the pumping
The corresponding data value range of each feature of the keyword and/or fuzzy word that take is consistent, formed the extraction keyword and/or
The characteristic of fuzzy word.
The creation module 34 creates special for the keyword of the extraction to the formation and/or the characteristic of fuzzy word
Levy data directory.
The display module 35, for being indexed according to the characteristic of the creation, search matches the wound in index data base
The data information for the characteristic index built, and the data information of the characteristic index for the matching creation that this is searched out
Search result is shown by relational degree taxis.
Optionally, the normalization module 33, can be specifically used for:
At least one feature is extracted from the keyword of the extraction and/or fuzzy word, at least one feature of the extraction
It is normalized, so that the corresponding data value range of each feature of the keyword of the extraction and/or fuzzy word is consistent,
Form the keyword of the extraction and/or the characteristic of fuzzy word.
Optionally, the display module 35, can be specifically used for:
It is indexed according to the characteristic of the creation and defines search result structure and rule, and the search result according to this definition
Structure and rule, search matches the data information of the characteristic index of the creation in index data base, and this is searched out
Matching the creation characteristic index data information search result by relational degree taxis displaying.
Fig. 4 is referred to, Fig. 4 is that the present invention is based on the structural representations of another embodiment of the ordering system of natural language analysis
Figure.It is different from an embodiment, the ordering system 40 based on natural language analysis described in the present embodiment further include: pushing module
41。
The pushing module 41, for pushing and should believe by the associated business of search result that relational degree taxis is shown to user
Breath.
Each unit module of the bright ordering system 30/40 based on natural language analysis can execute above method reality respectively
It applies and corresponds to step in example, therefore each unit module is not repeated herein, refer to the explanation of the above corresponding step.
Continuing with shown in Figure 5, Fig. 5 is the knot of the ordering system 3rd embodiment the present invention is based on natural language analysis
Structure schematic diagram.It is different from first embodiment, the ordering system 40 based on natural language analysis described in the present embodiment further include: particulate
Discriminance analysis module 43 is spent, to obtain more sophisticated category result.Specifically, normalizing network using posture in the present embodiment
To realize fine granularity discriminance analysis.Posture normalizes network for low-level image features such as conv5, fc6 for extracting by posture normalization
It is merged with unjustified fc8 advanced features, and uses the prediction position 2D DPM and 13 semantic position key points, Huo Zhezhi
It connects using the object frame and position markup information posture prototype having been provided, and then obtains more sophisticated category result.
The present invention provides a kind of sequencing equipment based on natural language analysis again, as shown in Figure 6, comprising: at least one
Manage device 51;And the memory 52 with the communication connection of at least one processor 51;Wherein, be stored with can be by least for memory 52
The instruction that one processor 51 executes, instruction is executed by least one processor 51, so that at least one processor 51 can be held
The above-mentioned sort method based on natural language analysis of row.
Wherein, memory 52 is connected with processor 51 using bus mode, and bus may include any number of interconnection
Bus and bridge, bus is by one or more processors 51 together with the various circuit connections of memory 52.Bus can also incite somebody to action
Together with various other circuit connections of management circuit or the like, these are all abilities for such as peripheral equipment, voltage-stablizer
Well known to domain, therefore, it will not be further described herein.Bus interface is provided between bus and transceiver and is connect
Mouthful.Transceiver can be an element, is also possible to multiple element, such as multiple receivers and transmitter, provides for passing
The unit communicated on defeated medium with various other devices.The data handled through processor 51 are carried out on the radio medium by antenna
Transmission, further, antenna also receives data and transfers data to processor 51.
Processor 51 is responsible for management bus and common processing, can also provide various functions, including timing, periphery connects
Mouthful, voltage adjusting, power management and other control functions.And memory 52 can be used for storage processor 51 and execute behaviour
Used data when making.
The present invention provides a kind of computer readable storage medium again, is stored with computer program.Computer program is processed
Device realizes above method embodiment when executing.
It can be found that above scheme, the available input information being input in search engine in search box, and using certainly
Right language analysis mode carries out keyword to the input information of the acquisition and/or fuzzy word extracts, and to the keyword of the extraction
And/or fuzzy word is normalized, so that the corresponding data of each feature of the keyword of the extraction and/or fuzzy word take
It is consistent to be worth range, forms the keyword of the extraction and/or the characteristic of fuzzy word, and the keyword of the extraction to the formation
And/or the characteristic creation characteristic index of fuzzy word, and indexed according to the characteristic of the creation, in index data
Search matches the data information of the characteristic index of the creation, and the characteristic for matching the creation that this is searched out in library
The search result of the data information of index is shown by relational degree taxis, be can be realized and is improved the search results ranking displaying degree of association.
Further, above scheme can extract at least one feature from the keyword of extraction and/or fuzzy word, right
At least one feature of the extraction is normalized, so that each feature pair of the keyword of the extraction and/or fuzzy word
The data value range answered is consistent, forms the keyword of the extraction and/or the characteristic of fuzzy word, such benefit is can
The calculation amount of keyword and/or fuzzy word is reduced, the search efficiency searched for based on keyword and/or fuzzy word is improved.
Further, above scheme can index according to the characteristic of creation and define search result structure and rule, and
According to the search result structure and rule of this definition, search matches the number of the characteristic index of the creation in index data base
It is believed that breath, and by the search result of the data information of the characteristic index for matching the creation searched out by relational degree taxis
It shows, such benefit is can to define different search result structure and rule according to the demand of user, so that is shown searches
Hitch fruit meets the different demands of user.
Further, above scheme can push and the associated quotient of search result by relational degree taxis displaying to user
Industry information can be realized the business for being input in search engine the input information association in search box with the user to user's push
The precision of information, the business information is high, needed for meeting user, can be improved the success rate of business promotion.
In several embodiments provided by the present invention, it should be understood that disclosed system, device and method can
To realize by another way.For example, device embodiments described above are only schematical, for example, module or
The division of unit, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units
Or component can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, institute
Display or the mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, device or unit
Indirect coupling or communication connection can be electrical property, mechanical or other forms.
Unit may or may not be physically separated as illustrated by the separation member, shown as a unit
Component may or may not be physical unit, it can and it is in one place, or may be distributed over multiple networks
On unit.It can select some or all of unit therein according to the actual needs to realize the mesh of present embodiment scheme
's.
In addition, each functional unit in each embodiment of the present invention can integrate in one processing unit, it can also
To be that each unit physically exists alone, can also be integrated in one unit with two or more units.It is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.
It, can if integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product
To be stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention substantially or
Say that all or part of the part that contributes to existing technology or the technical solution can embody in the form of software products
Out, which is stored in a storage medium, including some instructions are used so that a computer equipment
(can be personal computer, server or the network equipment etc.) or processor (processor) execute each implementation of the present invention
The all or part of the steps of methods.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. it is various
It can store the medium of program code.
The foregoing is merely section Examples of the invention, are not intended to limit protection scope of the present invention, all utilizations
Equivalent device made by description of the invention and accompanying drawing content or equivalent process transformation are applied directly or indirectly in other correlations
Technical field, be included within the scope of the present invention.
Claims (10)
1. a kind of sort method based on natural language analysis characterized by comprising
Obtain the input information being input in search engine in search box;
Using natural language analysis mode, keyword is carried out to the input information of the acquisition and/or fuzzy word extracts;
The keyword and/or fuzzy word of the extraction are normalized, so that the keyword of the extraction and/or fuzzy
The corresponding data value range of each feature of word is consistent, forms the keyword of the extraction and/or the characteristic of fuzzy word;
The characteristic creation characteristic index of keyword and/or fuzzy word to the extraction of the formation;
It is indexed according to the characteristic of the creation, search matches the characteristic index of the creation in index data base
Data information, and the search result of the data information of the characteristic index for the matching creation that described search is gone out is by association
Degree sequence is shown.
2. as described in claim 1 based on the sort method of natural language analysis, which is characterized in that described to the extraction
Keyword and/or fuzzy word are normalized, so that each feature of the keyword of the extraction and/or fuzzy word is corresponding
Data value range it is consistent, form the keyword of the extraction and/or the characteristic of fuzzy word, comprising:
At least one feature is extracted from the keyword of the extraction and/or fuzzy word, at least one feature of the extraction
It is normalized, so that the corresponding data value range one of each feature of the keyword of the extraction and/or fuzzy word
It causes, forms the keyword of the extraction and/or the characteristic of fuzzy word.
3. as described in claim 1 based on the sort method of natural language analysis, which is characterized in that described according to the creation
Characteristic index, search matches the data information of the characteristic index of the creation in index data base, and by institute
The search result for stating the data information of the characteristic index of the matching creation searched out is shown by relational degree taxis, is wrapped
It includes:
It is indexed according to the characteristic of the creation and defines search result structure and rule, and the search result according to the definition
Structure and rule, search matches the data information of the characteristic index of the creation in index data base, and searches described
The search result of the data information of the characteristic index for the matching creation that rope goes out is shown by relational degree taxis.
4. the sort method based on natural language analysis as described in claims 1 to 3 any one, which is characterized in that in institute
It states and is indexed according to the characteristic of the creation, search matches the number of the characteristic index of the creation in index data base
It is believed that breath, and the search result of the data information of the characteristic index for the matching creation that described search goes out is pressed into the degree of association
After sequence is shown, further includes:
To user's push and the associated business information of search result shown by relational degree taxis.
5. a kind of ordering system based on natural language analysis characterized by comprising
Obtain module, abstraction module, normalization module, creation module and display module;
The acquisition module, for obtaining the input information being input in search engine in search box;
The abstraction module, for using natural language analysis mode, to the input information of the acquisition carry out keyword and/or
Fuzzy word extracts;
The normalization module, for the extraction keyword and/or fuzzy word be normalized so that the pumping
The corresponding data value range of each feature of the keyword and/or fuzzy word that take is consistent, formed the extraction keyword and/
Or the characteristic of fuzzy word;
The creation module creates special for the keyword of the extraction to the formation and/or the characteristic of fuzzy word
Levy data directory;
The display module, for being indexed according to the characteristic of the creation, search matches the wound in index data base
The data information for the characteristic index built, and the data of the characteristic index for the matching creation that described search goes out are believed
The search result of breath is shown by relational degree taxis.
6. as claimed in claim 5 based on the ordering system of natural language analysis, which is characterized in that the normalization module,
It is specifically used for:
At least one feature is extracted from the keyword of the extraction and/or fuzzy word, at least one feature of the extraction
It is normalized, so that the corresponding data value range one of each feature of the keyword of the extraction and/or fuzzy word
It causes, forms the keyword of the extraction and/or the characteristic of fuzzy word.
7. as claimed in claim 5 based on the ordering system of natural language analysis, which is characterized in that the display module, tool
Body is used for:
It is indexed according to the characteristic of the creation and defines search result structure and rule, and the search result according to the definition
Structure and rule, search matches the data information of the characteristic index of the creation in index data base, and searches described
The search result of the data information of the characteristic index for the matching creation that rope goes out is shown by relational degree taxis.
8. the ordering system based on natural language analysis as described in claim 5 to 7 any one, which is characterized in that described
Ordering system based on natural language analysis, further includes:
Pushing module, for being pushed and the associated business information of search result shown by relational degree taxis to user.
9. a kind of sequencing equipment based on natural language analysis characterized by comprising
At least one processor;And
The memory being connect at least one described processor communication;Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
It manages device to execute, so that at least one described processor is able to carry out according to any one of claims 1 to 4 based on nature language
Say the sort method of analysis.
10. a kind of computer readable storage medium, is stored with computer program, which is characterized in that the computer program is located
Reason device realizes the sort method described in any one of Claims 1-4 based on natural language analysis when executing.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910331228.1A CN110059253A (en) | 2019-04-23 | 2019-04-23 | A kind of sort method and system and equipment based on natural language analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910331228.1A CN110059253A (en) | 2019-04-23 | 2019-04-23 | A kind of sort method and system and equipment based on natural language analysis |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110059253A true CN110059253A (en) | 2019-07-26 |
Family
ID=67320248
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910331228.1A Withdrawn CN110059253A (en) | 2019-04-23 | 2019-04-23 | A kind of sort method and system and equipment based on natural language analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110059253A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111028067A (en) * | 2019-12-28 | 2020-04-17 | 广东奥园奥买家电子商务有限公司 | E-commerce commodity searching method, device and equipment |
CN111062788A (en) * | 2019-12-28 | 2020-04-24 | 广东奥园奥买家电子商务有限公司 | E-commerce platform commodity recommendation method, device and equipment based on search |
CN113706260A (en) * | 2021-09-01 | 2021-11-26 | 镇江纵陌阡横信息科技有限公司 | E-commerce platform commodity recommendation method and device based on search content |
-
2019
- 2019-04-23 CN CN201910331228.1A patent/CN110059253A/en not_active Withdrawn
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111028067A (en) * | 2019-12-28 | 2020-04-17 | 广东奥园奥买家电子商务有限公司 | E-commerce commodity searching method, device and equipment |
CN111062788A (en) * | 2019-12-28 | 2020-04-24 | 广东奥园奥买家电子商务有限公司 | E-commerce platform commodity recommendation method, device and equipment based on search |
CN113706260A (en) * | 2021-09-01 | 2021-11-26 | 镇江纵陌阡横信息科技有限公司 | E-commerce platform commodity recommendation method and device based on search content |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5679993B2 (en) | Method and query system for executing a query | |
CN101876981B (en) | A kind of method and device building knowledge base | |
KR101339103B1 (en) | Document classifying system and method using semantic feature | |
CN102262765B (en) | Method and device for publishing commodity information | |
CN111143479A (en) | Knowledge graph relation extraction and REST service visualization fusion method based on DBSCAN clustering algorithm | |
US20160034514A1 (en) | Providing search results based on an identified user interest and relevance matching | |
US20130013616A1 (en) | Systems and Methods for Natural Language Searching of Structured Data | |
CN103123624B (en) | Determine method and device, searching method and the device of centre word | |
CN100462969C (en) | Method for providing and inquiry information for public by interconnection network | |
JP5616444B2 (en) | Method and system for document indexing and data querying | |
CN105493075A (en) | Retrieval of attribute values based upon identified entities | |
US7555428B1 (en) | System and method for identifying compounds through iterative analysis | |
CN105045852A (en) | Full-text search engine system for teaching resources | |
JP6165955B1 (en) | Method and system for matching images and content using whitelist and blacklist in response to search query | |
WO2010014082A1 (en) | Method and apparatus for relating datasets by using semantic vectors and keyword analyses | |
CN110059253A (en) | A kind of sort method and system and equipment based on natural language analysis | |
CN105843796A (en) | Microblog emotional tendency analysis method and device | |
CN108875065B (en) | Indonesia news webpage recommendation method based on content | |
CN113407785B (en) | Data processing method and system based on distributed storage system | |
CN109472008A (en) | A kind of Text similarity computing method, apparatus and electronic equipment | |
CN111400323A (en) | Data retrieval method, system, device and storage medium | |
CN111563382A (en) | Text information acquisition method and device, storage medium and computer equipment | |
US10650191B1 (en) | Document term extraction based on multiple metrics | |
CN105159927B (en) | Method and device for selecting subject term of target text and terminal | |
CN109992665A (en) | A kind of classification method based on the extension of problem target signature |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190726 |