CN110335114A - Classification method, device and the equipment of product - Google Patents
Classification method, device and the equipment of product Download PDFInfo
- Publication number
- CN110335114A CN110335114A CN201910579638.8A CN201910579638A CN110335114A CN 110335114 A CN110335114 A CN 110335114A CN 201910579638 A CN201910579638 A CN 201910579638A CN 110335114 A CN110335114 A CN 110335114A
- Authority
- CN
- China
- Prior art keywords
- product
- sorted
- keyword
- classification
- vocabulary
- 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
- 238000000034 method Methods 0.000 title claims abstract description 55
- 230000006870 function Effects 0.000 claims abstract description 273
- 238000013507 mapping Methods 0.000 claims abstract description 24
- 238000002372 labelling Methods 0.000 claims abstract description 15
- 230000001755 vocal effect Effects 0.000 claims abstract description 15
- 238000004891 communication Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 13
- 238000005065 mining Methods 0.000 claims description 11
- 239000000284 extract Substances 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 238000004519 manufacturing process Methods 0.000 description 11
- 230000000694 effects Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 5
- 230000011218 segmentation Effects 0.000 description 5
- 238000003058 natural language processing Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000008571 general function Effects 0.000 description 2
- 230000001105 regulatory effect Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 241001455214 Acinonyx jubatus Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000002045 lasting effect Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
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
- G06F16/35—Clustering; Classification
-
- 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
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0623—Item investigation
Abstract
Classification method, device and the equipment of a kind of product provided in an embodiment of the present invention.Wherein, a kind of classification method of product obtains the product information of the identical multiple products to be sorted of general utility functions classification；Wherein, the product information includes the verbal description to the featured function of the product to be sorted；From the product information of the multiple product to be sorted, keyword is extracted；The keyword is the vocabulary that can show that the featured function；For each product to be sorted, the featured function classification of the product to be sorted is determined using the mapping relations of preset keyword and featured function classification based on the keyword；The featured function classification of respectively each product labelling to be sorted product to be sorted.This programme may be implemented to classify to product according to the difference of featured function.
Description
Technical field
The present invention relates to behavior sorting technique fields, more particularly to a kind of classification method of product, device and equipment.
Background technique
Product relevant to internet, for example, the products such as virtual product and App (Application, application software) carry out
When analysis on competitive etc. is analyzed, in order to guarantee precision of analysis, Related product used in guaranteeing is generally required
Data are the data of the generic product of product to be analyzed；In the related technology, simple general function mostly just is carried out to product
The division of energy classification, for example, acquaintance's social category, stranger's social category, card class and Simulating management class etc..
But the main competitiveness of the product may be implemented in the featured function of Internet-related product, is to influence to be somebody's turn to do
The principal element of product competitiveness.For example, the featured function of a certain Video Applications is interactive can to determine development of action, it is another
The featured function of Video Applications is to possess most animation copyrights, and the featured function of a certain card class game application is that Dezhou is flutterred
Gram, the featured function of another card class game application is fried golden flower etc..In this regard, the above-mentioned difference according to general utility functions divides class
Otherwise, can only determine with product to be analyzed have similar general utility functions major class product, and can not determine with to point
Division product have the competing product of similar featured function.
Therefore, how to realize and classify according to the difference of featured function to product, be a problem to be solved.
Summary of the invention
The classification method for being designed to provide a kind of product, device and the equipment of the embodiment of the present invention improve void to realize
The effect of the convenience of the configuration inspection of quasi- article.Specific technical solution is as follows:
In a first aspect, the embodiment of the invention provides a kind of classification methods of product, this method comprises:
Obtain the product information of the identical multiple products to be sorted of general utility functions classification；Wherein, the product information includes
To the verbal description of the featured function of the product to be sorted；
From the product information of the multiple product to be sorted, keyword is extracted；The keyword is described to can show that
The vocabulary of featured function；
For each product to be sorted, it is based on the keyword, utilizes reflecting for preset keyword and featured function classification
Relationship is penetrated, determines the featured function classification of the product to be sorted；
The featured function classification of respectively each product labelling to be sorted product to be sorted.
Optionally, described from the product information of the multiple product to be sorted, extract keyword, comprising:
Based on the product information of the multiple product to be sorted, the first corpus is obtained；
The product information of the multiple product to be sorted is segmented, multiple words of the general utility functions classification are obtained
It converges；
Using first corpus and preset text mining mode, the important journey of the multiple vocabulary is calculated separately
Degree；
The specified quantity vocabulary that the significance level is met to predetermined keyword condition, is determined as the keyword.
Optionally, the product information based on the multiple product to be sorted obtains the first corpus, comprising:
Obtain the product information of multiple differentiated products；The general utility functions classification of the multiple differentiated products and it is the multiple to
The difference of sort product；
By in the product information of the multiple product to be sorted and the product information of the multiple differentiated products, general utility functions
The identical product information of classification obtains the first corpus comprising multiple corpus documents as a corpus document.
Optionally, the specified quantity vocabulary that the significance level is met to predetermined keyword condition, is determined as institute
State keyword, comprising:
The specified quantity vocabulary that the significance level is met to predetermined keyword condition, as candidate keywords；
For each candidate keywords, model is associated using preset search, obtains the candidate keywords in search engine
In associative information；
For each candidate keywords, whether judge in the associative information of the candidate keywords comprising for showing spy
The mark of color function；
If comprising the candidate keywords are determined as the keyword.
Optionally, it after the product information to the multiple product to be sorted segments, obtains described general
Before multiple vocabulary of functional category, the method also includes:
The product information of the multiple product to be sorted is segmented, multiple candidates of the general utility functions classification are obtained
Vocabulary；
The multiple vocabulary for obtaining the general utility functions classification, comprising:
Preset stop words is filtered from the multiple candidate vocabulary, obtains multiple vocabulary of the general utility functions classification；
The preset stop words is the intersection of the high frequency vocabulary in the product information of the product of each general utility functions classification；
And/or when in the multiple candidate vocabulary there are when english vocabulary, by meaning phase in the multiple candidate vocabulary
Vocabulary same, tense is different, is converted to the vocabulary of same tense, obtains multiple vocabulary of the general utility functions classification.
Optionally, it is directed to each product to be sorted described, is based on the keyword, utilizes preset keyword and characteristic
The mapping relations of functional category, after the step of determining the featured function classification of the product to be sorted, the method also includes:
Judgement is not determined by the quantity of the product to be sorted of the featured function classification, with the multiple product to be sorted
The ratio of total quantity, if meet preset proportion threshold value；
If conditions are not met, by the product to be sorted for being not determined by the featured function classification, as product to be sorted,
And return based on the keyword described in execution, using the mapping relations of preset keyword and featured function classification, determining should
The featured function classification of product to be sorted.
Second aspect, the embodiment of the invention provides a kind of sorter of product, which includes:
Obtaining product information module, for obtaining the product information of the identical multiple products to be sorted of general utility functions classification；
Wherein, the product information includes the verbal description to the featured function of the product to be sorted；
Keyword extracting module, for extracting keyword from the product information of the multiple product to be sorted；The pass
Keyword is the vocabulary that can show that the featured function；
Featured function category determination module is based on the keyword, utilization is preset for being directed to each product to be sorted
The mapping relations of keyword and featured function classification determine the featured function classification of the product to be sorted；
Featured function category label module, for being respectively the spy of each product labelling to be sorted product to be sorted
Color functional category.
Optionally, the keyword extracting module, comprising:
Corpus acquisition submodule obtains the first corpus for the product information based on the multiple product to be sorted；
Submodule is segmented, is segmented for the product information to the multiple product to be sorted, the general function is obtained
Multiple vocabulary of energy classification；
Significance level computational submodule is counted respectively for utilizing first corpus and preset text mining mode
Calculate the significance level of the multiple vocabulary；
Keyword determines submodule, for the significance level to be met to the specified quantity word of predetermined keyword condition
It converges, is determined as the keyword.
Optionally, the corpus acquisition submodule, is specifically used for:
Obtain the product information of multiple differentiated products；The general utility functions classification of the multiple differentiated products and it is the multiple to
The difference of sort product；
By in the product information of the multiple product to be sorted and the product information of the multiple differentiated products, general utility functions
The identical product information of classification obtains the first corpus comprising multiple corpus documents as a corpus document.
Optionally, the keyword determines submodule, is specifically used for:
The specified quantity vocabulary that the significance level is met to predetermined keyword condition, as candidate keywords；
For each candidate keywords, model is associated using preset search, obtains the candidate keywords in search engine
In associative information；
For each candidate keywords, whether judge in the associative information of the candidate keywords comprising for showing spy
The mark of color function；
If comprising the candidate keywords are determined as the keyword.
Optionally, the participle submodule, is specifically used for:
After the product information to the multiple product to be sorted segments, the more of the general utility functions classification are obtained
Before a vocabulary, the product information of the multiple product to be sorted is segmented, obtains the multiple of the general utility functions classification
Candidate vocabulary；
Preset stop words is filtered from the multiple candidate vocabulary, obtains multiple vocabulary of the general utility functions classification；
The preset stop words is the intersection of the high frequency vocabulary in the product information of the product of each general utility functions classification；
And/or when in the multiple candidate vocabulary there are when english vocabulary, by meaning phase in the multiple candidate vocabulary
Vocabulary same, tense is different, is converted to the vocabulary of same tense, obtains multiple vocabulary of the general utility functions classification.
Optionally, described device further include: judgment module, it is each for being directed in the featured function category determination module
Product to be sorted is based on the keyword, and using the mapping relations of preset keyword and featured function classification, determining should be wait divide
After the featured function classification of class product, judgement is not determined by the quantity of the product to be sorted of the featured function classification, with institute
State the ratio of the total quantity of multiple products to be sorted, if meet preset proportion threshold value；
If conditions are not met, trigger the featured function category determination module by be not determined by the featured function classification to
Sort product as product to be sorted, and returns described in execution based on the keyword, utilizes preset keyword and characteristic function
The mapping relations of energy classification, determine the featured function classification of the product to be sorted.
The third aspect, the embodiment of the invention provides a kind of electronic equipment, which includes:
Processor, communication interface, memory and communication bus, wherein processor, communication interface, memory pass through bus
Complete mutual communication；Memory, for storing computer program；Processor, for executing the journey stored on memory
Sequence, the step of realizing the classification method for the product that above-mentioned first aspect provides.
Fourth aspect is stored in the storage medium the embodiment of the invention provides a kind of computer readable storage medium
Computer program, the computer program realize the step of the classification method for the product that above-mentioned first aspect provides when being executed by processor
Suddenly.
In scheme provided in an embodiment of the present invention, due to the general utility functions classification that featured function is a certain product and the product
Under the different function of other products, also, product information includes treating the verbal description of the featured function of sort product, because
This, the product information of the identical multiple products to be sorted of available general utility functions classification, and on this basis can be from multiple
In the product information of product to be sorted, the keyword that can show that featured function is extracted；And then it is directed to each product to be sorted, base
In the keyword of the product to be sorted, using the mapping relations of preset keyword and featured function, it can determine that this is to be sorted
The featured function classification of product；To be featured function classification determined by the product labelling to be sorted, realize according to characteristic function
The difference of energy, the effect classified to product.As it can be seen that the difference according to featured function may be implemented to production by this programme
Product are classified.
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.
Fig. 1 is the flow diagram of the classification method for the product that one embodiment of the invention provides；
Fig. 2 be another embodiment of the present invention provides product classification method flow diagram；
Fig. 3 is the structural schematic diagram of the sorter for the product that one embodiment of the invention provides；
Fig. 4 be another embodiment of the present invention provides product sorter structural schematic diagram；
Fig. 5 is the structural schematic diagram for the electronic equipment that one embodiment of the invention provides.
Specific embodiment
In order to make those skilled in the art more fully understand the technical solution in the present invention, implement below in conjunction with the present invention
Attached drawing in example, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment
Only a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, the common skill in this field
Art personnel every other embodiment obtained without creative efforts belongs to the model that the present invention protects
It encloses.
The classification method of the product of one embodiment of the invention is introduced first below.
The classification method of product provided in an embodiment of the present invention, can be applied to electronic equipment, which specifically may be used
To include desktop computer, portable computer, internet television, intelligent mobile terminal, server and wearable intelligence
Terminal etc., is not limited thereto, and any electronic equipment that the embodiment of the present invention may be implemented belongs to the guarantor of the embodiment of the present invention
Protect range.
As shown in Figure 1, the process of the classification method of the product of one embodiment of the invention, this method may include walking as follows
It is rapid:
S101 obtains the product information of the identical multiple products to be sorted of general utility functions classification；Wherein, product information includes
Treat the verbal description of the featured function of sort product.
In a particular application, in order to attract user, often believed using the product of the featured function of the verbal description product
Breath, also, featured function is a certain product function different from other products under the general utility functions classification of the product.Therefore,
The product information of the identical multiple products to be sorted of available general utility functions classification, with for by subsequent step S102 extremely
S104 realization treats sort product by featured function and classifies.
Wherein, the mode for obtaining the product information of the identical multiple products to be sorted of general utility functions classification, specifically can be
A variety of.Illustratively, when the product market for showing or selling product, such as in application shop, there are multiple productions to be sorted
When the general utility functions classification of product, the product information of all products to be sorted under a certain general utility functions classification can be directly acquired.
For example, obtaining the product information of all social applications in a certain application shop.Alternatively, it is illustrative, it is produced when showing or selling
It, can be from the production of multiple products to be sorted there is no when the general utility functions classification of multiple products to be sorted in the product market of product
The vocabulary for showing general utility functions classification is extracted in product information；It will indicate that the identical product to be sorted of the vocabulary of general utility functions classification,
It is determined as the identical product of general utility functions classification.
The mode of any product information for obtaining the identical multiple products to be sorted of general utility functions classification, is used equally for this hair
It is bright, the present embodiment to this with no restriction.
S102 extracts keyword from the product information of multiple products to be sorted；Keyword is that can show that featured function
Vocabulary.
In a particular application, from the product information of multiple products to be sorted, the method for extracting keyword specifically be can be
A variety of.Illustratively, the text for describing featured function usually has well-regulated expression format, so that keyword is in product information
The middle well-regulated position of tool, for example, " this product has the advantage that ' keyword ' ", " bright spot is ' keyword ' " and " especially
, ' keyword ' " etc. expression format.Therefore, it can use regular expression, from the product information of each product to be sorted,
Extract keyword.Alternatively, it is illustrative, it can use natural language processing technique, from the product information of each product to be sorted,
Extract keyword.In order to facilitate understanding and rational deployment, subsequent to be extracted in Fig. 2 embodiment to using natural language processing technique
The method of keyword, is specifically described.
In addition, it is a variety of to can show that the keyword of featured function specifically can be.For example, it can be that " Dezhou is flutterred
Gram ", " fried golden flower ", " plot interaction " and " copyright quantity is most " etc. vocabulary.
S103 is based on keyword for each product to be sorted, utilizes reflecting for preset keyword and featured function classification
Relationship is penetrated, determines the featured function classification of the product to be sorted.
In a particular application, for each product to be sorted, it is based on keyword, utilizes preset keyword and featured function
The mapping relations of classification determine the featured function classification of the product to be sorted, can specifically include: being directed to each production to be sorted
Product match the keyword extracted with the product information to be sorted, if successful match, utilize preset key
The keyword of the successful match is mapped as featured function classification, mapping is obtained by the mapping relations of word and featured function classification
Featured function classification, the featured function classification as the product to be sorted.
Wherein, the mapping relations of preset keyword and featured function classification specifically can be a variety of.Illustratively,
For each product to be sorted, when the quantity of the keyword of successful match is 1, which can be with are as follows: by keyword
As featured function classification.Alternatively, it is illustrative, for each product to be sorted, when the quantity of the keyword of successful match is
When multiple, which can be with are as follows: using the most keyword of successful match number as featured function classification.Alternatively, example
Property, when the quantity of identified keyword is multiple, which can be with are as follows: utilizes the first corpus and preset text
This mining model calculates the significance level of each keyword, and after sorting from high to low by significance level, specified quantity is closed before taking
Keyword is as featured function classification.Wherein, the step of the first corpus and preset text mining mode and Fig. 2 embodiment of the present invention
It is identical in rapid S204.
S104, the featured function classification of respectively each product labelling to be sorted product to be sorted.
It in a particular application, can be in the featured function class for determining the product to be sorted for each product to be sorted
When other, for featured function classification determined by the product labelling to be sorted.Wherein, label featured function class specifically may be used otherwise
To be a variety of.Illustratively, product to be sorted can be stored in the storage region for belonging to the featured function classification.Alternatively,
Featured function classification is added in the product information of product to be sorted, as class label.
In scheme provided in an embodiment of the present invention, due to the general utility functions classification that featured function is a certain product and the product
Under the different function of other products, also, product information includes treating the verbal description of the featured function of sort product, because
This, the product information of the identical multiple products to be sorted of available general utility functions classification, and on this basis can be from multiple
In the product information of product to be sorted, the keyword that can show that featured function is extracted；And then it is directed to each product to be sorted, base
In the keyword of the product to be sorted, using the mapping relations of preset keyword and featured function, it can determine that this is to be sorted
The featured function classification of product；To be featured function classification determined by the product labelling to be sorted, realize according to characteristic function
The difference of energy, the effect classified to product.As it can be seen that the difference according to featured function may be implemented to production by this programme
Product are classified.
As shown in Fig. 2, the process of the classification method of the product of another embodiment of the present invention, this method may include walking as follows
It is rapid:
S201 obtains the product information of the identical multiple products to be sorted of general utility functions classification；Wherein, product information includes
Treat the verbal description of the featured function of sort product.
S201 is identical step with the S101 of Fig. 1 embodiment of the present invention, and details are not described herein, and it is real to be detailed in Fig. 1 of the present invention
Apply the description of example.
S202 obtains the first corpus based on the product information of multiple products to be sorted.
In a particular application, the product information based on multiple products to be sorted obtains the mode of the first corpus, specifically may be used
To be a variety of.It is specifically described in the form of alternative embodiment below.
In a kind of optional embodiment, above-mentioned steps S202 be can specifically include: by the production of multiple products to be sorted
Product information, as the first corpus.
The general utility functions classification of multiple products to be sorted is identical, and a certain vocabulary is believed in the integral product of multiple products to be sorted
The frequency of appearance in breath, the frequency are able to reflect " the characteristic degree " of the vocabulary, and the frequency is higher, " the characteristic journey of the vocabulary
Degree " is lower, related to the significance level of the vocabulary.It therefore, can be by the product information of multiple products to be sorted, as the first language
Library is expected, so that the important journey of each vocabulary in product information can be calculated using the first corpus in subsequent step S204
Degree, for extracting keyword.
In another optional embodiment, above-mentioned steps S202 can specifically include following steps:
Obtain the product information of multiple differentiated products；The general utility functions classification of multiple differentiated products and multiple products to be sorted
Difference；
By in the product information of multiple products to be sorted and the product information of multiple differentiated products, general utility functions classification is identical
Product information obtain the first corpus comprising multiple corpus documents as a corpus document.
Wherein, multiple differentiated products can be a variety of.Illustratively, multiple differentiated products can be it is multiple with it is to be sorted
The general utility functions classification of product, and the identical product of general utility functions classification of each differentiated products.Alternatively, it is illustrative, it is multiple
Differentiated products can be multiple and product to be sorted general utility functions classification, and in each differentiated products, at least one difference produces
Product are different from the general utility functions of other differentiated products.Illustratively, the general utility functions classification of product to be sorted is " game ", multiple
Differentiated products can be multiple general utility functions classifications be " education ", " music ", and/or, " social activity " etc. and general utility functions class
Not " game " different product.
Also, it is retouched in the acquisition modes of the product information of multiple differentiated products, with the step S101 of Fig. 1 embodiment of the present invention
The acquisition modes of the product information for the multiple products to be sorted stated are similar, and difference is to obtain the product information of multiple differentiated products
When, it needs to be obtained for the general utility functions classification different from the general utility functions classification of multiple products to be sorted.For identical
Details are not described herein for part.
Product information and multiple products to be sorted in the above-mentioned another optional embodiment, using differentiated products
Product information obtains the first corpus.It, and directly will be multiple at this point, have differences the product information of product in the first corpus
The product information of product to be sorted is compared as the first corpus, can be improved the significance level of keyword and non-key word
Difference degree when reducing the significance level of keyword and non-key word and being closer to, the erroneous judgement that is easy to appear and/or is failed to judge, is mentioned
The accuracy that high keyword determines.
S203 segments the product information of multiple products to be sorted, obtains multiple vocabulary of general utility functions classification.
In a particular application, the mode segmented to the product information of multiple products to be sorted, specifically can be a variety of
's.Illustratively, when product information is English, due between english vocabulary there are space, space symbol can be made
For the mark of vocabulary segmentation, vocabulary segmentation is carried out to product information, obtains multiple vocabulary.Alternatively, it is illustrative, work as product information
When for Chinese, Chinese word segmentation tool can be used.Wherein, Chinese word segmentation tool can be a variety of, for example, Chinese word segmentation
It can be jieba (stammerer) participle, language cloud participle and BosonNLP participle etc..
Also, after the product information to multiple products to be sorted segments, the multiple of general utility functions classification are obtained
The mode of vocabulary specifically can be a variety of.It is illustrated in the form of alternative embodiment below.
In a kind of optional embodiment, the above-mentioned product information to multiple products to be sorted is segmented, and is led to
With multiple vocabulary of functional category, it can specifically include following steps:
The product information of multiple products to be sorted is segmented, obtains multiple vocabulary, and obtained multiple vocabulary are made
For multiple vocabulary of the general utility functions classification of multiple products to be sorted.
At this point, since the general utility functions classification of multiple products to be sorted is identical, to the product of multiple products to be sorted
Information is segmented, and obtained multiple vocabulary are equivalent to multiple vocabulary of the general utility functions classification.Therefore, it can will segment
Multiple vocabulary of the multiple vocabulary arrived as the general utility functions classification of multiple products to be sorted.
In another optional embodiment, the above-mentioned product information to the multiple product to be sorted is segmented,
Multiple vocabulary of the general utility functions classification are obtained, can specifically include following steps:
The product information of multiple products to be sorted is segmented, multiple candidate vocabulary of general utility functions classification are obtained；
Preset stop words is filtered from multiple candidate vocabulary, obtains multiple vocabulary of general utility functions classification；It is preset to stop
Word is the intersection of the high frequency vocabulary in the product information of the product of each general utility functions classification；
And/or when in multiple candidate vocabulary there are when english vocabulary, the identical, tense by meaning in multiple candidate vocabulary
Different vocabulary is converted to the vocabulary of same tense, obtains multiple vocabulary of general utility functions classification.
Wherein, the setting of preset stop words can be a variety of.Illustratively, can directly will according to the language habits
Preset stop words is set as without semantic word.For example, it can set default for preposition and modal particle etc.
Stop words.Alternatively, it is illustrative, the product information of the product of different general utility functions classifications can be segmented respectively,
Obtain multiple vocabulary of every kind of general utility functions classification；Word frequency is selected to be greater than from multiple vocabulary of every kind of general utility functions classification default
The vocabulary of word frequency threshold obtains high frequency lexical set；The intersection for taking the high frequency lexical set of each general utility functions classification, as pre-
If stop words.The set-up mode of any preset stop words is used equally for the present invention, the present embodiment to this with no restriction.
Also, by the vocabulary that meaning is identical, tense is different in multiple candidate vocabulary, be converted to the side of the vocabulary of same tense
Formula specifically can be a variety of.It is exemplary, the vocabulary that meaning is identical, tense is different in multiple candidate vocabulary can be replaced with
Specified vocabulary.Wherein, any one of tense vocabulary when specified vocabulary can be no tense, past tense and carry out etc..Example
Such as, by the candidate vocabulary " played " of candidate vocabulary " playing " and past tense when carrying out, the vocabulary of no tense is replaced with
"play".Alternatively, it is illustrative, meaning can be changed to by by the vocabulary that meaning is identical, tense is different in multiple candidate vocabulary
In vocabulary identical, tense is different, the most vocabulary of frequency of occurrence.For example, " playing " frequency of occurrence is 3 times,
" played " frequency of occurrence is 1 time, " played " can be replaced with " playing ".
And illustrate after being converted to same vocabulary, the statistics number of the different vocabulary of script tense is included in same vocabulary.It reduces
The problem of keyword extraction inaccuracy caused by the significance level of different tense dispersion vocabulary.
In above-mentioned another alternative embodiment, after being segmented to the product information of multiple products to be sorted, obtain
The multiple vocabulary arrived pre-process candidate vocabulary as candidate vocabulary: stop words filtering, and/or, morphology turns
It changes, so that pretreated candidate vocabulary will be passed through as multiple vocabulary of general utility functions classification.Participle is obtained with directly more
A vocabulary is compared as multiple vocabulary of the general utility functions classification of multiple products to be sorted, it is possible to reduce is preset without semantic
Influence of the stop words to the significance level of vocabulary, multiple vocabulary of the general utility functions classification guaranteed are with semantic word
It converges, improves the accuracy that keyword determines.Also, it, can be by the statistics of the different vocabulary of script tense after carrying out morphology conversion
Number is included in same vocabulary, to reduce different tenses but the identical multiple vocabulary of meaning, disperses the important journey of the meaning vocabulary
After degree, caused by keyword extraction inaccuracy problem.
S204 calculates separately the significance level of multiple vocabulary using the first corpus and preset text mining mode.
In a particular application, using the first corpus and preset text mining mode, the weight of multiple vocabulary is calculated separately
Degree is wanted, specifically can be a variety of.It illustratively, can when the first corpus is the product information of multiple products to be sorted
To be directed to each vocabulary, the reverse document-frequency of reflection vocabulary frequency of occurrences in the first corpus is calculated, and reflection should
The word frequency of vocabulary frequency of occurrence in the product information for obtaining the vocabulary；For each vocabulary, by the reverse file frequency of the vocabulary
Rate is multiplied with text word frequency, obtains the significance level of the vocabulary.Alternatively, it is illustrative, when the first corpus is to include multiple languages
Expect document corpus when, can be directed to each vocabulary, calculate reflect the vocabulary in the first corpus the frequency of occurrences it is reverse
Document-frequency, and reflect the vocabulary in the text word frequency with frequency of occurrence in the corpus document comprising vocabulary；For each word
It converges, the reverse document-frequency of the vocabulary is multiplied with text word frequency, obtains the significance level of the vocabulary.
Wherein, reverse document-frequency specifically can be the total number of corpus document in the first corpus, divided by including the word
The quantity of the corpus document of remittance, and take logarithm to obtain obtained quotient.Text word frequency specifically can be the vocabulary and go out in the text
Existing number is obtained divided by the summation of vocabulary frequency of occurrence all in the text.
It is any to utilize the first corpus and preset text mining mode, calculate separately the significance level of multiple vocabulary
Method, be used equally for the present invention, the present embodiment to this with no restriction.
Significance level is met the specified quantity vocabulary of predetermined keyword condition, is determined as keyword by S205.
In a particular application, predetermined keyword condition specifically can be a variety of.Illustratively, predetermined keyword condition can
Significance level to be vocabulary is greater than preset significance level threshold value, alternatively, according to vocabulary volume significance level from big to small suitable
Sequence sequence, the vocabulary of specified quantity before coming.Any predetermined keyword condition, is used equally for the present invention, the present embodiment is to this
With no restriction.
Also, the specified quantity vocabulary that significance level is met to predetermined keyword condition, is determined as the mode of keyword,
It specifically can be a variety of.It is specifically described in the form of alternative embodiment below.
In a kind of optional embodiment, above-mentioned steps S205 may include: that significance level is met predetermined keyword
The specified quantity vocabulary of condition, as keyword.
When the significance level of vocabulary meets predetermined keyword condition, show the vocabulary in the production for describing featured function
In product information, it is more likely that be the vocabulary that can directly show featured function, therefore, can directly meet significance level default
The specified quantity vocabulary of keyword condition, as keyword.
In another optional embodiment, above-mentioned steps S205 can specifically include following steps:
The specified quantity vocabulary that significance level is met to predetermined keyword condition, as candidate keywords；
For each candidate keywords, model is associated using preset search, obtains the candidate keywords in search engine
In associative information；
For each candidate keywords, whether judge in the associative information of the candidate keywords comprising for showing characteristic function
The mark of energy；
If comprising the candidate keywords are determined as keyword.
Wherein, preset search association model is the relevant data of data in search engine, for obtaining to being inputted
Model, specifically can be a variety of.Illustratively, preset search association model can be the search connection of google search engine
Think API (Application Programming Interface, application programming interfaces), the search association of cheetah search engine
API, alternatively, the search of search engine must be answered to associate API.
Also, it is a variety of for showing that the mark of featured function can be in associative information.Illustratively, which can be with
It is the vocabulary with " featured function " meaning.For example, it can be " game, rule (game, playing method) ",
" Characteristic function (featured function) " or " strong point (bright spot) " etc. vocabulary.Alternatively, example
Property, which can be the specified sign for representing featured function data, for example, when this is identified as " 1 " or "comprising" etc.
When specified sign, show the data in the associative information of candidate keywords comprising showing featured function.When judging result is candidate
When in the associative information of keyword including the mark for showing featured function, show that the candidate keywords are that can show that characteristic
Therefore the candidate keywords can be determined as keyword by the vocabulary of function.
In above-mentioned another alternative embodiment, significance level is met to the specified quantity word of predetermined keyword condition
It converges, verifies whether each candidate keywords are about characteristic function as candidate keywords, and using preset search association model
The vocabulary of energy, thus by keyword is determined as by the candidate keywords verified.Significance level is met with directly default crucial
The specified quantity vocabulary of entry part, is determined as keyword and compares, by search association to the finger for meeting predetermined keyword condition
Fixed number amount vocabulary is verified, and the accuracy of the keyword as the vocabulary for showing featured function can be improved.
S206 is based on keyword for each product to be sorted, utilizes reflecting for preset keyword and featured function classification
Relationship is penetrated, determines the featured function classification of the product to be sorted.
S207, the featured function classification of respectively each product labelling to be sorted product to be sorted.
S206 to S207 is identical step with the S103 to S104 of Fig. 1 embodiment of the present invention, and details are not described herein, is detailed in
The description of Fig. 1 embodiment of the present invention.
In above-mentioned Fig. 2 embodiment, keyword is extracted using natural language processing technique: using the first corpus and being preset
Text mining mode, the significance level of multiple vocabulary under the general utility functions classification of product to be sorted is calculated, thus by important
Degree meets the specified quantity vocabulary of predetermined keyword condition, is determined as keyword.It is non-natural with utilization regular expression etc.
Language processing techniques carry out keyword extraction and compare, and can preferably cope with diversified product information description form, Ke Yibao
Demonstrate,proving extracted keyword is the vocabulary for reflecting product information semanteme, to improve the accuracy of keyword extraction.
Optionally, in the step S103 of aforementioned present invention Fig. 1 embodiment, or the step S206 of Fig. 2 embodiment of the present invention
Later, the classification method of product provided in an embodiment of the present invention can also include the following steps:
Judgement is not determined by the quantity of the product to be sorted of featured function classification, with the total quantity of multiple products to be sorted
Ratio, if meet preset proportion threshold value；
If conditions are not met, the product to be sorted that will be not determined by featured function classification, as product to be sorted, and returns and holds
Row determines the characteristic of the product to be sorted using the mapping relations of preset keyword and featured function classification based on keyword
Functional category.
Wherein, the sum when the quantity for the product to be sorted for being not determined by featured function classification, with multiple products to be sorted
The ratio of amount, when being unsatisfactory for preset proportion threshold value, it is opposite for showing that this is not determined by the product to be sorted of featured function classification
For minority product, can show that the vocabulary of the featured function of minority's product goes out in the product information of multiple applications to be sorted
Existing number is in contrast less, is not extracted by as keyword, therefore minority's product is not determined by featured function classification.
At this point, needing to be divided using minority's product as product to be sorted to guarantee that minority's product can be classified
Class.It is equivalent to from multiple products to be sorted, eliminates the in contrast more public product that classification is completed.Therefore, may be used
Influence to avoid the vocabulary in the product information of mass product to vocabulary in the product information of minority's product, to realize to spy
The determination of color the function in contrast keyword extraction and featured function classification of minority's product of minority.
Also, it will be not determined by the product to be sorted of featured function classification, as product to be sorted, and return to execution and be based on
Keyword determines the featured function class of the product to be sorted using the mapping relations of preset keyword and featured function classification
Other specific method, similar with the step in Fig. 1 and Fig. 2 embodiment of the present invention, difference is that featured function class will be not determined by
Other product to be sorted, the object of product to be sorted are minority's game.For same section, details are not described herein, is detailed in this hair
Bright Fig. 1 and Fig. 2 embodiment, and the description of corresponding alternative embodiment.
In this alternative embodiment, can by the way that the product to be sorted of featured function classification will be not determined by, as to point
Class product, and return to execution based on keyword, using the mapping relations of preset keyword and featured function classification, determine this to
The featured function classification of sort product realizes that minority's product to featured function in contrast minority is classified, to improve
The success rate of product classification.Also, the quantity to the product to be sorted for being not determined by featured function classification, with multiple productions to be sorted
The ratio of the total quantity of product, is made whether to meet the judgement of preset proportion threshold value, will be not determined by featured function class with lasting
Other product to be sorted carries out classification as product to be sorted and compares, and is conducive to filter in contrast excessively minority, market value
Possible little product reduces classification cost.Certainly, preset proportion threshold value can be set by user according to self-demand
It sets, to meet user to the different demands of classification success rate.For example, it can be set to 20% or 10% etc..
Corresponding to above method embodiment, one embodiment of the invention additionally provides the sorter of product.
As shown in figure 3, the sorter for the product that one embodiment of the invention provides, the apparatus may include:
Obtaining product information module 301, the product for obtaining the identical multiple products to be sorted of general utility functions classification are believed
Breath；Wherein, the product information includes the verbal description to the featured function of the product to be sorted；
Keyword extracting module 302, for extracting keyword from the product information of the multiple product to be sorted；Institute
Stating keyword is the vocabulary that can show that the featured function；
Featured function category determination module 303 is based on the keyword, using pre- for being directed to each product to be sorted
If keyword and featured function classification mapping relations, determine the featured function classification of the product to be sorted；
Featured function category label module 304, for being respectively the institute of each product labelling to be sorted product to be sorted
State featured function classification.
In scheme provided in an embodiment of the present invention, due to the general utility functions classification that featured function is a certain product and the product
Under the different function of other products, also, product information includes treating the verbal description of the featured function of sort product, because
This, the product information of the identical multiple products to be sorted of available general utility functions classification, and on this basis can be from multiple
In the product information of product to be sorted, the keyword that can show that featured function is extracted；And then it is directed to each product to be sorted, base
In the keyword of the product to be sorted, using the mapping relations of preset keyword and featured function, it can determine that this is to be sorted
The featured function classification of product；To be featured function classification determined by the product labelling to be sorted, realize according to characteristic function
The difference of energy, the effect classified to product.As it can be seen that the difference according to featured function may be implemented to production by this programme
Product are classified.
As shown in figure 4, another embodiment of the present invention provides product sorter, the apparatus may include:
Obtaining product information module 401, the product for obtaining the identical multiple products to be sorted of general utility functions classification are believed
Breath；Wherein, the product information includes the verbal description to the featured function of the product to be sorted；
Keyword extracting module 402, comprising:
Corpus acquisition submodule 4021 obtains the first language for the product information based on the multiple product to be sorted
Expect library；
Submodule 4022 is segmented, is segmented for the product information to the multiple product to be sorted, is obtained described logical
With multiple vocabulary of functional category；
Significance level computational submodule 4023, for utilizing first corpus and preset text mining mode, point
The significance level of the multiple vocabulary is not calculated；
Keyword determines submodule 4024, for the significance level to be met to the specified quantity of predetermined keyword condition
Vocabulary is determined as the keyword.
Featured function category determination module 403 is based on the keyword, using pre- for being directed to each product to be sorted
If keyword and featured function classification mapping relations, determine the featured function classification of the product to be sorted；
Featured function category label module 404, for being respectively the institute of each product labelling to be sorted product to be sorted
State featured function classification.
Optionally, the corpus acquisition submodule 4021, is specifically used for:
Obtain the product information of multiple differentiated products；The general utility functions classification of the multiple differentiated products and it is the multiple to
The difference of sort product；
By in the product information of the multiple product to be sorted and the product information of the multiple differentiated products, general utility functions
The identical product information of classification obtains the first corpus comprising multiple corpus documents as a corpus document.
Optionally, the keyword determines submodule 4024, is specifically used for:
The specified quantity vocabulary that the significance level is met to predetermined keyword condition, as candidate keywords；
For each candidate keywords, model is associated using preset search, obtains the candidate keywords in search engine
In associative information；
For each candidate keywords, whether judge in the associative information of the candidate keywords comprising for showing spy
The mark of color function；
If comprising the candidate keywords are determined as the keyword.
Optionally, the participle submodule 4022, is specifically used for:
After the product information to the multiple product to be sorted segments, the more of the general utility functions classification are obtained
Before a vocabulary, the product information of the multiple product to be sorted is segmented, obtains the multiple of the general utility functions classification
Candidate vocabulary；
Preset stop words is filtered from the multiple candidate vocabulary, obtains multiple vocabulary of the general utility functions classification；
The preset stop words is the intersection of the high frequency vocabulary in the product information of the product of each general utility functions classification；
And/or when in the multiple candidate vocabulary there are when english vocabulary, by meaning phase in the multiple candidate vocabulary
Vocabulary same, tense is different, is converted to the vocabulary of same tense, obtains multiple vocabulary of the general utility functions classification.
Optionally, described device further include: judgment module is used in the featured function category determination module 303, or
Featured function category determination module 403 is directed to each product to be sorted, is based on the keyword, utilizes preset keyword and spy
The mapping relations of color functional category, after the featured function classification for determining the product to be sorted, judgement is not determined by the characteristic
The quantity of the product to be sorted of functional category, the ratio with the total quantity of the multiple product to be sorted, if meet preset
Proportion threshold value；
If conditions are not met, triggering the featured function category determination module 303 or featured function category determination module
403 will be not determined by the product to be sorted of the featured function classification, and as product to be sorted, and it is described based on institute to return to execution
Keyword is stated, using the mapping relations of preset keyword and featured function classification, determines the featured function of the product to be sorted
Classification.
Corresponding to above-described embodiment, the embodiment of the invention also provides a kind of electronic equipment, as shown in figure 5, the equipment can
To include:
Processor 501, communication interface 502, memory 503 and communication bus 504, wherein processor 501, communication interface
502, memory logical 503 crosses communication bus 504 and completes mutual communication；
Memory 503, for storing computer program；
Processor 501 when for executing the computer program stored on above-mentioned memory 503, realizes above-described embodiment
In any product classification method the step of.
In scheme provided in an embodiment of the present invention, due to the general utility functions classification that featured function is a certain product and the product
Under the different function of other products, also, product information includes treating the verbal description of the featured function of sort product, because
This, the product information of the identical multiple products to be sorted of available general utility functions classification, and on this basis can be from multiple
In the product information of product to be sorted, the keyword that can show that featured function is extracted；And then it is directed to each product to be sorted, base
In the keyword of the product to be sorted, using the mapping relations of preset keyword and featured function, it can determine that this is to be sorted
The featured function classification of product；To be featured function classification determined by the product labelling to be sorted, realize according to characteristic function
The difference of energy, the effect classified to product.As it can be seen that the difference according to featured function may be implemented to production by this programme
Product are classified.
Above-mentioned memory may include RAM (Random Access Memory, random access memory), also may include
NVM (Non-Volatile Memory, nonvolatile memory), for example, at least a magnetic disk storage.Optionally, memory
It can also be that at least one is located away from the storage device of above-mentioned processor.
Above-mentioned processor can be general processor, including CPU (Central Processing Unit, central processing
Device), NP (Network Processor, network processing unit) etc.；Can also be DSP (Digital Signal Processor,
Digital signal processor), ASIC (Application Specific Integrated Circuit, specific integrated circuit),
FPGA (Field-Programmable Gate Array, field programmable gate array) or other programmable logic device are divided
Vertical door or transistor logic, discrete hardware components.
The computer readable storage medium that one embodiment of the invention provides, is contained in electronic equipment, this is computer-readable to deposit
It is stored with computer program in storage media, when which is executed by processor, realizes any product in above-described embodiment
Classification method the step of.
In scheme provided in an embodiment of the present invention, due to the general utility functions classification that featured function is a certain product and the product
Under the different function of other products, also, product information includes treating the verbal description of the featured function of sort product, because
This, the product information of the identical multiple products to be sorted of available general utility functions classification, and on this basis can be from multiple
In the product information of product to be sorted, the keyword that can show that featured function is extracted；And then it is directed to each product to be sorted, base
In the keyword of the product to be sorted, using the mapping relations of preset keyword and featured function, it can determine that this is to be sorted
The featured function classification of product；To be featured function classification determined by the product labelling to be sorted, realize according to characteristic function
The difference of energy, the effect classified to product.As it can be seen that the difference according to featured function may be implemented to production by this programme
Product are classified.
In another embodiment provided by the invention, a kind of computer program product comprising instruction is additionally provided, when it
When running on computers, so that computer executes the classification method of any product in above-described embodiment.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When implemented in software, it can entirely or partly realize in the form of a computer program product.The computer program
Product includes one or more computer instructions.When loading on computers and executing the computer program instructions, all or
It partly generates according to process or function described in the embodiment of the present invention.The computer can be general purpose computer, dedicated meter
Calculation machine, computer network or other programmable devices.The computer instruction can store in computer readable storage medium
In, or from a computer readable storage medium to the transmission of another computer readable storage medium, for example, the computer
Instruction can pass through wired (such as coaxial cable, optical fiber, DSL from a web-site, computer, server or data center
(Digital Subscriber Line, digital operation maintenance personnel line) or wireless (such as: infrared ray, radio, microwave etc.) mode
It is transmitted to another web-site, computer, server or data center.The computer readable storage medium can be
Any usable medium that computer can access either includes the integrated server of one or more usable mediums, data center
Equal data storage devices.The usable medium can be magnetic medium, (for example, floppy disk, hard disk, tape), optical medium (such as:
DVD (Digital Versatile Disc, digital versatile disc)) or semiconductor medium (such as: SSD (Solid State
Disk, solid state hard disk)) etc..
Herein, relational terms such as first and second and the like be used merely to by an entity or operation with it is another
One entity or operation distinguish, and without necessarily requiring or implying between these entities or operation, there are any this reality
Relationship or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to the packet of nonexcludability
Contain, so that the process, method, article or equipment for including a series of elements not only includes those elements, but also including
Other elements that are not explicitly listed, or further include for elements inherent to such a process, method, article, or device.
In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including the element
Process, method, article or equipment in there is also other identical elements.
Each embodiment in this specification is all made of relevant mode and describes, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device and
For electronic equipment embodiment, since it is substantially similar to the method embodiment, so be described relatively simple, related place referring to
The part of embodiment of the method illustrates.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modification, equivalent replacement, improvement and so within the spirit and principles in the present invention, are all contained in protection scope of the present invention
It is interior.
Claims (10)
1. a kind of classification method of product, which is characterized in that the described method includes:
Obtain the product information of the identical multiple products to be sorted of general utility functions classification；Wherein, the product information includes to institute
State the verbal description of the featured function of product to be sorted；
From the product information of the multiple product to be sorted, keyword is extracted；The keyword is that can show that the characteristic
The vocabulary of function；
For each product to be sorted, it is based on the keyword, is closed using the mapping of preset keyword and featured function classification
System, determines the featured function classification of the product to be sorted；
The featured function classification of respectively each product labelling to be sorted product to be sorted.
2. the method according to claim 1, wherein the product information from the multiple product to be sorted
In, extract keyword, comprising:
Based on the product information of the multiple product to be sorted, the first corpus is obtained；
The product information of the multiple product to be sorted is segmented, multiple vocabulary of the general utility functions classification are obtained；
Using first corpus and preset text mining mode, the significance level of the multiple vocabulary is calculated separately；
The specified quantity vocabulary that the significance level is met to predetermined keyword condition, is determined as the keyword.
3. according to the method described in claim 2, it is characterized in that, the product letter based on the multiple product to be sorted
Breath obtains the first corpus, comprising:
Obtain the product information of multiple differentiated products；The general utility functions classification of the multiple differentiated products with it is the multiple to be sorted
The difference of product；
By in the product information of the multiple product to be sorted and the product information of the multiple differentiated products, general utility functions classification
Identical product information obtains the first corpus comprising multiple corpus documents as a corpus document.
4. according to the method described in claim 2, it is characterized in that, described meet predetermined keyword condition for the significance level
Specified quantity vocabulary, be determined as the keyword, comprising:
The specified quantity vocabulary that the significance level is met to predetermined keyword condition, as candidate keywords；
For each candidate keywords, model is associated using preset search, obtain the candidate keywords in a search engine
Associative information；
For each candidate keywords, whether judge in the associative information of the candidate keywords comprising for showing characteristic function
The mark of energy；
If comprising the candidate keywords are determined as the keyword.
5. a kind of sorter of product, which is characterized in that described device includes:
Obtaining product information module, for obtaining the product information of the identical multiple products to be sorted of general utility functions classification；Wherein,
The product information includes the verbal description to the featured function of the product to be sorted；
Keyword extracting module, for extracting keyword from the product information of the multiple product to be sorted；The keyword
For the vocabulary that can show that the featured function；
Featured function category determination module is based on the keyword, utilizes preset key for being directed to each product to be sorted
The mapping relations of word and featured function classification determine the featured function classification of the product to be sorted；
Featured function category label module, for being respectively the characteristic function of each product labelling to be sorted product to be sorted
It can classification.
6. device according to claim 5, which is characterized in that the keyword extracting module, comprising:
Corpus acquisition submodule obtains the first corpus for the product information based on the multiple product to be sorted；
Submodule is segmented, is segmented for the product information to the multiple product to be sorted, obtains the general utility functions class
Other multiple vocabulary；
Significance level computational submodule calculates separately institute for utilizing first corpus and preset text mining mode
State the significance level of multiple vocabulary；
Keyword determines submodule, for the significance level to be met to the specified quantity vocabulary of predetermined keyword condition, really
It is set to the keyword.
7. device according to claim 6, which is characterized in that the corpus acquisition submodule is specifically used for:
Obtain the product information of multiple differentiated products；The general utility functions classification of the multiple differentiated products with it is the multiple to be sorted
The difference of product；
By in the product information of the multiple product to be sorted and the product information of the multiple differentiated products, general utility functions classification
Identical product information obtains the first corpus comprising multiple corpus documents as a corpus document.
8. device according to claim 6, which is characterized in that the keyword determines submodule, is specifically used for:
The specified quantity vocabulary that the significance level is met to predetermined keyword condition, as candidate keywords；
For each candidate keywords, model is associated using preset search, obtain the candidate keywords in a search engine
Associative information；
For each candidate keywords, whether judge in the associative information of the candidate keywords comprising for showing characteristic function
The mark of energy；
If comprising the candidate keywords are determined as the keyword.
9. a kind of electronic equipment, which is characterized in that including processor, communication interface, memory and communication bus, wherein processing
Device, communication interface, memory complete mutual communication by bus；Memory, for storing computer program；Processor,
For executing the program stored on memory, the method and step as described in claim 1-4 is any is realized.
10. a kind of computer readable storage medium, which is characterized in that computer program is stored in the storage medium, it is described
The method and step as described in claim 1-4 is any is realized when computer program is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910579638.8A CN110335114A (en) | 2019-06-28 | 2019-06-28 | Classification method, device and the equipment of product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910579638.8A CN110335114A (en) | 2019-06-28 | 2019-06-28 | Classification method, device and the equipment of product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110335114A true CN110335114A (en) | 2019-10-15 |
Family
ID=68144667
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910579638.8A Pending CN110335114A (en) | 2019-06-28 | 2019-06-28 | Classification method, device and the equipment of product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110335114A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111949763A (en) * | 2020-07-15 | 2020-11-17 | 有米科技股份有限公司 | Document material-based method and device for determining categories of single articles |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345421A (en) * | 2013-06-20 | 2013-10-09 | 华为技术有限公司 | Method and device for pushing application program |
| CN103577423A (en) * | 2012-07-23 | 2014-02-12 | 阿里巴巴集团控股有限公司 | Keyword classification method and system |
| US9229621B2 (en) * | 2012-05-22 | 2016-01-05 | Paletteapp, Inc. | Electronic palette system |
| CN105354194A (en) * | 2014-08-19 | 2016-02-24 | 上海中怡通信息科技有限公司 | Intelligent commodity classifying method and system |
| CN106649455A (en) * | 2016-09-24 | 2017-05-10 | 孙燕群 | Big data development standardized systematic classification and command set system |
| CN107169021A (en) * | 2017-04-07 | 2017-09-15 | 华为机器有限公司 | Method and apparatus for predicting application function label |
| CN107247780A (en) * | 2017-06-12 | 2017-10-13 | 北京理工大学 | A kind of patent document method for measuring similarity of knowledge based body |
| CN107809526A (en) * | 2017-09-28 | 2018-03-16 | 努比亚技术有限公司 | End application sorting technique, mobile terminal and computer-readable recording medium |
| CN107992633A (en) * | 2018-01-09 | 2018-05-04 | 国网福建省电力有限公司 | Electronic document automatic classification method and system based on keyword feature |
| CN108228546A (en) * | 2018-01-19 | 2018-06-29 | 北京中关村科金技术有限公司 | A kind of text feature, device, equipment and readable storage medium storing program for executing |
| US20180189615A1 (en) * | 2017-01-03 | 2018-07-05 | Samsung Electronics Co., Ltd. | Electronic apparatus and method of operating the same |
-
2019
- 2019-06-28 CN CN201910579638.8A patent/CN110335114A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9229621B2 (en) * | 2012-05-22 | 2016-01-05 | Paletteapp, Inc. | Electronic palette system |
| CN103577423A (en) * | 2012-07-23 | 2014-02-12 | 阿里巴巴集团控股有限公司 | Keyword classification method and system |
| CN103345421A (en) * | 2013-06-20 | 2013-10-09 | 华为技术有限公司 | Method and device for pushing application program |
| CN105354194A (en) * | 2014-08-19 | 2016-02-24 | 上海中怡通信息科技有限公司 | Intelligent commodity classifying method and system |
| CN106649455A (en) * | 2016-09-24 | 2017-05-10 | 孙燕群 | Big data development standardized systematic classification and command set system |
| US20180189615A1 (en) * | 2017-01-03 | 2018-07-05 | Samsung Electronics Co., Ltd. | Electronic apparatus and method of operating the same |
| CN107169021A (en) * | 2017-04-07 | 2017-09-15 | 华为机器有限公司 | Method and apparatus for predicting application function label |
| CN107247780A (en) * | 2017-06-12 | 2017-10-13 | 北京理工大学 | A kind of patent document method for measuring similarity of knowledge based body |
| CN107809526A (en) * | 2017-09-28 | 2018-03-16 | 努比亚技术有限公司 | End application sorting technique, mobile terminal and computer-readable recording medium |
| CN107992633A (en) * | 2018-01-09 | 2018-05-04 | 国网福建省电力有限公司 | Electronic document automatic classification method and system based on keyword feature |
| CN108228546A (en) * | 2018-01-19 | 2018-06-29 | 北京中关村科金技术有限公司 | A kind of text feature, device, equipment and readable storage medium storing program for executing |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111949763A (en) * | 2020-07-15 | 2020-11-17 | 有米科技股份有限公司 | Document material-based method and device for determining categories of single articles |
| CN111949763B (en) * | 2020-07-15 | 2024-03-22 | 有米科技股份有限公司 | Single-item category determining method and device based on text material |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110020422B (en) | Feature word determining method and device and server | |
| CN108241741B (en) | Text classification method, server and computer readable storage medium | |
| CN107992477B (en) | Text subject determines method and device | |
| CN108322473A (en) | User behavior analysis method and apparatus | |
| CN108304375A (en) | A kind of information identifying method and its equipment, storage medium, terminal | |
| CN107784092A (en) | A kind of method, server and computer-readable medium for recommending hot word | |
| US20130060769A1 (en) | System and method for identifying social media interactions | |
| US11361030B2 (en) | Positive/negative facet identification in similar documents to search context | |
| CN103425710A (en) | Subject-based searching method and device | |
| US10740377B2 (en) | Identifying categories within textual data | |
| CN110263248A (en) | A kind of information-pushing method, device, storage medium and server | |
| US10546088B2 (en) | Document implementation tool for PCB refinement | |
| US10956470B2 (en) | Facet-based query refinement based on multiple query interpretations | |
| US20140188456A1 (en) | Dictionary Markup System and Method | |
| CN110390044A (en) | A kind of searching method and equipment of the similar network page | |
| WO2012158572A2 (en) | Exploiting query click logs for domain detection in spoken language understanding | |
| CN109472008A (en) | A kind of Text similarity computing method, apparatus and electronic equipment | |
| CN114416998A (en) | Text label identification method and device, electronic equipment and storage medium | |
| CN109492401B (en) | Content carrier risk detection method, device, equipment and medium | |
| CN109325135A (en) | Text based video generation method, device, computer equipment and storage medium | |
| CN107273546A (en) | Counterfeit application detection method and system | |
| CN110516062A (en) | A kind of search processing method and device of document | |
| CN110245357A (en) | Principal recognition methods and device | |
| CN111814481B (en) | Shopping intention recognition method, device, terminal equipment and storage medium | |
| US20210271637A1 (en) | Creating descriptors for business analytics applications |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210527 Address after: 25, 5th floor, shuangjingfang office building, 3 frisha street, Singapore Applicant after: Zhuomi Private Ltd. Address before: Room 1101, Santai Commercial Building, 139 Connaught Road, Hong Kong, China Applicant before: HONG KONG LIVE.ME Corp.,Ltd. |
|
| AD01 | Patent right deemed abandoned | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20221104 |