CN103377249B - Keyword put-on method and system - Google Patents
Keyword put-on method and system Download PDFInfo
- Publication number
- CN103377249B CN103377249B CN201210129565.0A CN201210129565A CN103377249B CN 103377249 B CN103377249 B CN 103377249B CN 201210129565 A CN201210129565 A CN 201210129565A CN 103377249 B CN103377249 B CN 103377249B
- Authority
- CN
- China
- Prior art keywords
- keyword
- key element
- dictionary
- search
- word
- 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.)
- Active
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of keyword put-on method, comprise the following steps：Step S1, obtain keyword to be put；Step S2, calculate the transition probability of the keyword；Step S3, keyword that transition probability reaches to a first threshold launched by SEM channels, the keyword that conversion ratio is not up to the first threshold is launched by SEO channels.The invention also discloses a kind of keyword jettison system.The present invention can select out the keyword that suitable e-commerce website is launched automatically, and suitable dispensing channel is selected according to the transition probability of each keyword, taken into account keyword dispensing cost and keyword launch caused by income, realize keyword dispensing efficient management.
Description
Technical field
The present invention relates to a kind of keyword put-on method and system, more particularly to a kind of automatic dispensing e-commerce website
Search engine marketing keyword keyword put-on method and system.
Background technology
Today's society, internet have been deep among life, and search engine is convenient and swift with its, and information content is huge etc.
Advantage receives the favor of user, turns into the entrance of numerous user's online.Ecommerce is also made in fast development in recent years
Stand in great numbers into businessman, situation with keen competition.E-commerce venture will show one's talent in competition, do search engine marketing well,
User is allowed accurately to find out corresponding product in a search engine most important.
In traditional search engine marketing, keyword management mode be operated with manpower manual based on, sent out in ecommerce
The early stage of exhibition, range of goods is few, and business Relative vertical, labor management can meet business demand.But nowadays, electronic business
Business company all develops to comprehensive shopping platform, and product variety is up to hundreds thousand of, the side of labor management search engine keywords
Formula is unable to do what one wishes, turns into the bottleneck in search engine marketing.
The inferior position of manpower management SEO (search engine optimization)/SEM (search engine marketing) keyword has：
1st, enough keywords can not be found.Calculated within 1 minute with 1 word, 1 staff can only launch 100,000 in 1 year
Keyword, and the keyword demand of large-scale electric business is millions of.
2nd, keyword is chaotic, it is difficult to forms institutional framework.Because manpower can not accurately remember the keyword once launched,
Also it is difficult to the species of accurate division keyword, can causes to repeat to launch, meaning of a word confusion is unclear.
3rd, it is difficult to the value for judging keyword.The bad word of a large amount of changing effects can be produced by launching by hand, be caused greatly
The wasting of resources.
For electronic business enterprise keyword demand it is big the characteristics of, it is necessary to a set of supermatic, intelligent
Algorithm come ensure SEO/SEM launch keyword quantity and quality.
The content of the invention
The technical problem to be solved in the present invention is for labor management when overcoming that keyword demand is larger in the prior art
Keyword causes that keyword management is chaotic, easily causes the defects of launching transition probability relatively low keyword, there is provided a kind of automatic
Management keyword, the keyword put-on method and system be easy to keyword classification, can interpolate that keyword value.
The present invention is that solve above-mentioned technical problem by following technical proposals：
A kind of keyword put-on method, its feature are that it comprises the following steps：
Step S1, obtain keyword to be put；
Step S2, calculate the transition probability of the keyword, transition probability represents that user has been searched for after the keyword, bought
The probability of the commodity related to the keyword, transition probability can be expressed as：(purchase number/search results pages number of clicks) *
100%；
Step S3, keyword that transition probability reaches to a first threshold launched by SEM channels.
Preferably, step S3In it is further comprising the steps of：The keyword that conversion ratio is not up to the first threshold passes through SEO
Channel is launched.I.e. the higher keyword of transition probability can emphasis launch, and the relatively low keyword of transition probability can not throw charge
SEM channels, and based on free SEO.
Preferably, step S1Comprise the following steps：
Step S11, obtain product information from a product database, and according to the product information determine product category with
And dictionary corresponding with the product category is called according to the product category；
Step S12, the product information is divided at least one key element；
Step S13, using the dictionary and according to each key element by product information cutting word at least one key element keyword, with
And at least one key element keyword is formatted with the form of unified key element keyword；
Step S14, calculate an investment yield of each key element keyword, and select investment yield to reach one second threshold
The key element keyword of value is as the keyword.The calculation formula of wherein investment yield is：ROI (investment yield)=Sale
(search in Website sales volume)/(Search (search in Website number) * CPC (average clicking cost)).It is obvious that not all production
Product information in product database is suitable for being launched, in order to improve the income of keyword dispensing, when selecting key element keyword,
Main selection investment yield reaches the key element keyword of certain setting value (Second Threshold) as keyword to be put, with flat
The income that weighing apparatus launches the expenditure of keyword and keyword is brought.
Preferably, step S13In in a manner of unifying synonym using a synonym dictionary, using the side of unified capital and small letter
Formula and the one or more in the mode for replacing half double byte character are used to format at least one key element keyword with unified
The form of key element keyword.Such as the unification of synonym：" hair dryer " and " hair-dryer "；Or wrong word corrigendum：" Nuo Ji
It is sub- " and " promise note is sub- "；Or Chinese and English matching is unified：The mode such as " Dell " and " dell " unifies the key element keyword
Form.
Preferably, the dictionary be brand dictionary, generic name of the goods dictionary, unit dictionary, attribute dictionary, model dictionary and
One or more in conventional word dictionary.Conventional word dictionary is the dictionary for for example containing all conventional preposition and auxiliary word.
Preferably, step S12Afterwards, step S13It is before further comprising the steps of：Using conventional word dictionary excision product letter
Preposition and/or auxiliary word in breath.
Preferably, at least one key element is selected from following key element：Brand key element, generic name of the goods key element, unit key element,
Attribute key element and model key element.
Preferably, step S14In comprise the following steps：
Step S141, filter out the key element keyword of the unit key element, the attribute key element and the model key element；
Step S142, calculate through step S141Investment yield of remaining each key element keyword after filtering out, and select
Investment yield reaches the key element keyword of a Second Threshold as the keyword.Such as the unit key element such as " 3KG ", " 500 grams "
Key element keyword be not worth, be generally rejected.
Preferably, step S1Comprise the following steps：
Step S11, from a search history database obtain search key, have recorded use in the search history database
The search key of family search and the search rate of each search key；
Step S12, filter out a part in the search key, the part is：Preposition, auxiliary word, unit, attribute and model；
Step S13, will be through step S12Search key after filtering out is as the keyword.
The search history database includes the word search key that user searches in search engine, user is in ecommerce
Search key scanned in website etc., this kind of word can accurately match the search custom of user, can obtain excellent
The chance first shown.
Preferably, step S3Comprise the following steps：
Step S31, judge whether the transition probability of each keyword reaches the first threshold, if so, into step S32；If
It is no, into step S33；
Step S32, the keyword launched by SEM channels；
Step S33, judge whether the search rate of the keyword reaches one the 3rd threshold value, if so, into step S34；If
It is no, into step S35；
Step S34, the keyword launched by SEO channels；
Step S35, abandon the dispensing of the keyword.
I.e. although some transition probabilities are not high but the keyword that is frequently searched, and with certain launching value, can
To be launched by free SEO channels.
The present invention also provides a kind of keyword jettison system, and its feature is that it includes：
One keyword acquisition device, for obtaining keyword to be put；
One computing device, for calculating the transition probability of the keyword；
One keyword delivery device, the keyword for transition probability to be reached to a first threshold are launched by SEM channels.
Preferably, the keyword delivery device is additionally operable to conversion ratio being not up to the keyword of the first threshold and passes through SEO
Channel is launched.
Preferably, the keyword acquisition device includes：
One information acquisition device, for obtaining product information from a product database, and it is true according to the product information
Determine product category and dictionary corresponding with the product category is called according to the product category；
One key element generating means, for the product information to be divided into at least one key element；
One cutting word device, at least one key element closing product information cutting word using the dictionary and according to each key element
Keyword, and at least one key element keyword is formatted with the form of unified key element keyword；
One keyword generating means, for calculating an investment yield of each key element keyword, and select investment return
Rate reaches the key element keyword of a Second Threshold as the keyword.
Preferably, the cutting word device is additionally operable in a manner of unifying synonym using a synonym dictionary, using unified big
The mode of small letter and one or more in replacing by the way of half double byte character are crucial to format at least one key element
Word is with the form of unified key element keyword.
Preferably, the dictionary be brand dictionary, generic name of the goods dictionary, unit dictionary, attribute dictionary, model dictionary and
One or more in conventional word dictionary.
Preferably, the cutting word device is additionally operable to using the preposition and/or auxiliary word in conventional word dictionary excision product information.
Preferably, at least one key element is selected from following key element：Brand key element, generic name of the goods key element, unit key element,
Attribute key element and model key element.
Preferably, the keyword generating means also include：
One filters out unit, for filtering out the key element keyword of the unit key element, the attribute key element and the model key element；
One computing unit, for calculating the investment yield of remaining each key element keyword after filtering out, and select
Investment yield reaches the key element keyword of a Second Threshold as the keyword.
Preferably, the keyword acquisition device includes：
One search information acquisition device, for obtaining search key from a search history database, the search history
The search key of user's search and the search rate of each search key are have recorded in database；
One filtering device, for filtering out the part in the search key, the part is：Preposition, auxiliary word, unit, category
Property and model；
One keyword generating means, for using the search key after the filtering device filters out as the keyword.
Preferably, the keyword delivery device includes：
One first judgment means, for judging whether the transition probability of each keyword reaches the first threshold, if so, opening
With one first delivery device；If it is not, enable one second judgment means；
First delivery device, for the keyword to be launched by SEM channels；
Second judgment means, whether the search rate for judging the keyword reaches one the 3rd threshold value, if so, enabling
One second delivery device；If it is not, abandon the dispensing of the keyword；
Second delivery device, for the keyword to be launched by SEO channels.
The positive effect of the present invention is：, can be certainly the invention provides a kind of keyword put-on method and system
The dynamic keyword selected out suitable e-commerce website and launched, and it is suitable to select according to the transition probability of each keyword
Launch channel, taken into account keyword dispensing cost and keyword launch caused by income, realize keyword dispensing height
Effect management.
Brief description of the drawings
Fig. 1 is the flow chart of the keyword put-on method of the embodiment of the present invention 1.
Fig. 2 is the structured flowchart of the keyword jettison system of the embodiment of the present invention 1.
Fig. 3 is the flow chart of the keyword put-on method of the embodiment of the present invention 2.
Fig. 4 is the structured flowchart of the keyword jettison system of the embodiment of the present invention 2.
Embodiment
Present pre-ferred embodiments are provided below in conjunction with the accompanying drawings, to describe technical scheme in detail.
Embodiment 1
With reference to figure 1, the keyword put-on method described in the present embodiment, comprise the following steps：
Step 101, obtain product information from a product database, and according to the product information determine product category with
And dictionary corresponding with the product category is called according to the product category, the product database can be carried by e-commerce website
The product database of the e-commerce website sale supplied, the dictionary are brand dictionary, generic name of the goods dictionary, unit word
One or more in storehouse, attribute dictionary, model dictionary and conventional word dictionary.
Step 102, the product information is divided at least one key element, at least one key element is selected from following key element：Brand
Key element, generic name of the goods key element, unit key element, attribute key element and model key element.
Step 103, the preposition and/or auxiliary word in product information are cut off using conventional word dictionary.
Step 104, using the dictionary and according to each key element by product information cutting word at least one key element keyword, with
And at least one key element keyword is formatted with the form of unified key element keyword.It is wherein unified same using a synonym dictionary
The mode of adopted word, the one or more by the way of unified capital and small letter and in replacing by the way of half double byte character are come form
Change at least one key element keyword with the form of unified key element keyword.
Step 105, an investment yield of each key element keyword is calculated, and selects investment yield to reach one second threshold
The key element keyword of value is as the keyword.Specifically, the unit key element, the attribute key element and the model key element are first filtered out
Key element keyword, such as the key element keyword " 5kg " of unit key element, the key element keyword " red " of attribute key element and model will
Key element keyword " N97 " of element etc., the investment yield of remaining each key element keyword after filtering out then is being calculated, and
Selection investment yield reaches the key element keyword of a Second Threshold as the keyword.
Step 106, the transition probability of the keyword is calculated, transition probability is expressed as：(purchase number/search results pages point
Hit number) * 100%.
Step 107, judge whether the transition probability of the keyword reaches first threshold, if so, into step 108；If it is not,
Into step 109.
Step 108, the keyword is launched by SEM channels.
Step 109, the keyword is launched by SEO channels.
With reference to figure 2, the keyword jettison system described in the present embodiment is introduced, it includes：
One keyword acquisition device 1, for obtaining keyword to be put；
One computing device 2, for calculating the transition probability of the keyword；
One keyword delivery device 3, the keyword for transition probability to be reached to a first threshold are thrown by SEM channels
Put, the keyword that conversion ratio is not up to the first threshold is launched by SEO channels.
Specifically, the keyword acquisition device 1 includes：
One information acquisition device 11, for obtaining product information from a product database, and according to the product information
Determine product category and dictionary corresponding with the product category is called according to the product category, wherein, the dictionary is brand word
One or more in storehouse, generic name of the goods dictionary, unit dictionary, attribute dictionary, model dictionary and conventional word dictionary；
One key element generating means 12, for the product information to be divided into at least one key element, wherein, at least one key element
Selected from following key element：Brand key element, generic name of the goods key element, unit key element, attribute key element and model key element；
One cutting word device 13, for using the dictionary and according to each key element by product information cutting word at least one key element
Keyword, and at least one key element keyword is formatted with the form of unified key element keyword；
One keyword generating means 14, for calculating an investment yield of each key element keyword, and investment is selected to receive
Beneficial rate reaches the key element keyword of a Second Threshold as the keyword.
Wherein, the cutting word device 13 is additionally operable to by the way of a synonym dictionary unifies synonym, using unified size
The mode write and the one or more in replacing by the way of half double byte character format at least one key element keyword
With the form of unified key element keyword.And the cutting word device 13 is additionally operable to using in conventional word dictionary excision product information
Preposition and/or auxiliary word.
More specifically, the keyword generating means 14 also include：
One filters out unit 141, for filtering out the key element keyword of the unit key element, the attribute key element and the model key element；
One computing unit 142, for calculating the investment yield of remaining each key element keyword after filtering out, and select
Select investment yield and reach the key element keyword of a Second Threshold as the keyword.
Embodiment 2
With reference to figure 3, the keyword put-on method described in the present embodiment is introduced, it comprises the following steps：
Step 101, search key is obtained from a search history database, use is have recorded in the search history database
The search key of family search and the search rate of each search key.
Step 102, the part in the search key is filtered out, the part is：Preposition, auxiliary word, unit, attribute and type
Number.
Step 103, the search key after being filtered out through step, 102 is as the keyword.
Step 104, the transition probability of the keyword is calculated.
Step 105, judge whether the transition probability of each keyword reaches the first threshold, if so, into step 106；
If it is not, into step 107.
Step 106, the keyword is launched by SEM channels.
Step, 107, judge whether the search rate of the keyword reaches one the 3rd threshold value, if so, into step 108；If
It is no, into step 109.
Step 108, the keyword is launched by SEO channels.
Step 109, the dispensing of the keyword is abandoned.
With reference to figure 4, the keyword jettison system described in the present embodiment is introduced, it includes：
One keyword acquisition device 1, for obtaining keyword to be put；
One computing device 2, for calculating the transition probability of the keyword；
One keyword delivery device 3, the keyword for transition probability to be reached to a first threshold are thrown by SEM channels
Put, the keyword that conversion ratio is not up to the first threshold is launched by SEO channels.
As different from Example 1, the keyword acquisition device 1 includes：
One search information acquisition device 11, for obtaining search key from a search history database, the search is gone through
The search key of user's search and the search rate of each search key are have recorded in history database；
One filtering device 12, for filtering out the part in the search key, the part is：Preposition, auxiliary word, unit,
Attribute and model；
One keyword generating means 13, for using the search key after the filtering device filters out as the keyword.
Wherein, the keyword delivery device 3 includes：
One first judgment means 31, for judging whether the transition probability of each keyword reaches the first threshold, if so,
Enable one first delivery device 32；If it is not, enable one second judgment means 33；
First delivery device 32, for the keyword to be launched by SEM channels；
Second judgment means 33, whether the search rate for judging the keyword reaches one the 3rd threshold value, if so, opening
With one second delivery device 34；If it is not, abandon the dispensing of the keyword；
Second delivery device 34, for the keyword to be launched by SEO channels.
Embodiment 3
The principle of embodiment 3 is with embodiment 1 and all same of embodiment 2, difference：
The acquisition source of keyword to be put had both included the product database described in embodiment 1, also including the institute of embodiment 2
The search history database stated.
Below with an application example, technical scheme is further described.
First, keyword jettison system of the invention automatically takes out the suitable electric business and thrown by being docked with electric quotient data storehouse
All keywords put (have expense and cost, if the keyword of dispensing can not be brought " certain value " due to launching keyword
Sale return, then it is considered as the keyword of " improper " dispensing.The standard of " certain value " is different because of company, such as can be first this
Standard is set to the keyword that keyword of the rate of return on investment more than 500% is exactly " being adapted to launch ", that is to say, that launches crucial
Word, every 1 yuan of cost, can bring 5 yuan of sales volumes, can be regarded as suitable dispensing).Taking word source can be：
Word source one is taken, sells product word：The product information of electric business sale is obtained, such as the product information is carried from electric business
Obtained in the database of confession, the keyword that so just can guarantee that dispensing is all related to the electric business, and skill is segmented using Chinese character
Section text information is grown in art, cutting, obtains being adapted to the words such as the product brand, product classification, product information of dispensing and filters out not
Unit of the product information needed etc., cutting word filtering mainly judge according to part of speech, and brand word, classificating word are considered as big valuable of flow
The word of value, product information word flow is relatively small, but conversion ratio is higher, and " 3KG ", " 500 grams " this specification word are not worth,
Generally it is rejected.The word remained by subsequent analysis, it is necessary to prejudge conversion ratio.Such as：ProductName " profoundness automatically contains
Gold spin fragrance elite upgrading dress liquid detergent 3kg " will be split for：" profoundness " (brand), " liquid detergent " (classification), " it is complete from
It is dynamic ", " gold is spun ", " fragrance " (product information).
Take word source two, user's search term：Including user search engine search term and user searching in electric business station
Rope word, this kind of word can accurately match the search custom of user, the chance preferentially shown can be obtained, for this kind of search
Word, cutting word is not required to, but needs to filter, for example, user often searches some nonsensical words, such as " 11111 ", " 2 yuan "
These words need to be filtered out with the filter method in above cutting word.
By taking word technology, the degree of association of dispensing keyword and electric business product to be guaranteed automatically, because electric business
There is provided brand, classification, product information database, then word is taken based on database, brand word, the product word for ensureing to launch all are these
What electric business was sold, speed can be greatly improved again, lift productivity.
Then, analysis takes keyword obtained by word step, and to historic user behavioral data, (the historic user behavioral data is for example
Above-mentioned user's search term, after mainly having user's search, purchase probability, commodity category total sales volume, goods orders average amount, search
User's volumes of searches in engine) carrying out data mining, (it is a variety of numbers that the user behavior data of electric business will be collected arrangement by the system
According to all data update daily, and data are for example shown in table 1), it can be deduced that the transition probability of each word, the high word of transition probability
Meeting emphasis is launched, and conversion ratio is low, but the word more than occurrence number can not throw the SEM channels of charge, and based on free SEO.
We can sell state with the inventory status of analysis of key word corresponding product, avoid dispensing from selling commodity and cause to waste.
The historic user behavioral data list of table 1
With reference to Chinese search technology, can analyze each word in station search result quality (because technical strength is different,
The quality good or not of electric business search in Website result is uneven, and low-quality search result is such as：Search for " rice ", out former results are rice
Clothes, the furniture of color, rather than rice.This low-quality search result and user's request gap are too remote, under conversion ratio, no
It is adapted to launch), the high word of quality will use search page, and for landing page, quality is low, can go to product detail page or advertising campaign
Page.
By key word analysis technology, good guarantee will be obtained by launching the quality of keyword so that and dispensing shoots the arrow at the target,
Reduce and waste.
Then, begin to launch the keyword automatically.By the keyword for screening and analyzing, it will according to the commodity of electric business
The automatic classification of classification, search engine (such as being docked by search engine opening API) is published to by being docked with search engine.Mesh
Directory structures and electric business be consistent (for example, electric business commodity classification bibliographic structure, in SEM, search engine provides several layers of catalogues,
Facilitate user's sort key word.It is assumed that the corresponding SEM " dispensing account " of electric business one-level classification " food and drink ", two level classification " import
The corresponding SEM " popularization plan " of food ", the corresponding SEM " popularization unit " of electric business three-level classification " import milk ", cutting word in import milk
The suitable dispensing keyword obtained, it will launch to below its corresponding popularization unit.Advantage of this is that electricity can be combined
Popularization strategy of the business to different major classes, emphasis major class, which increases, is launched, and non-emphasis, which is reduced, to be launched) such that management keyword more has from now on
Orderliness.
Certain described above be only for example, the keyword put-on method and system of the present invention to be not limited to this several
Kind implementation.
Although the foregoing describing the embodiment of the present invention, it will be appreciated by those of skill in the art that these
It is merely illustrative of, protection scope of the present invention is defined by the appended claims.Those skilled in the art is not carrying on the back
On the premise of principle and essence from the present invention, various changes or modifications can be made to these embodiments, but these are changed
Protection scope of the present invention is each fallen within modification.
Claims (20)
1. a kind of keyword put-on method, it is characterised in that it comprises the following steps：
Step S1, obtain keyword to be put；
Step S2, calculate the transition probability of the keyword；Transition probability represents that user is searched for after the keyword, purchase and the pass
The probability of the related commodity of keyword, the calculation formula of transition probability are：(purchase number/search results pages number of clicks) *
100%；
Step S3, keyword that transition probability reaches to a first threshold launched by SEM channels.
2. the keyword put-on method as described in claim l, it is characterised in that step S3In it is further comprising the steps of：Will conversion
The keyword that rate is not up to the first threshold is launched by SEO channels.
3. the keyword put-on method as described in claim l, it is characterised in that step S1Comprise the following steps：
Step S11, product information is obtained from a product database, and product category and root are determined according to the product information
Dictionary corresponding with the product category is called according to the product category；
Step S12, the product information is divided at least one key element；
Step S13, using the dictionary and according to each key element by product information cutting word at least one key element keyword, to pass
Formulaization at least one key element keyword is with the form of unified key element keyword；
Step S14, calculate an investment yield of each key element keyword, and select investment yield to reach a Second Threshold
Key element keyword is as the keyword.
4. keyword put-on method as claimed in claim 3, it is characterised in that step S13In with using a synonym dictionary unite
The mode of one synonym, the one or more by the way of unified capital and small letter and in replacing by the way of half double byte character come
At least one key element keyword is formatted with the form of unified key element keyword.
5. keyword put-on method as claimed in claim 3, it is characterised in that the dictionary is brand dictionary, commodity common name
Claim the one or more in dictionary, unit dictionary, attribute dictionary, model dictionary and conventional word dictionary.
6. keyword put-on method as claimed in claim 5, it is characterised in that step S12Afterwards, step S13Also include before
Following steps：Preposition and/or auxiliary word in product information is cut off using conventional word dictionary.
7. keyword put-on method as claimed in claim 3, it is characterised in that at least one key element is selected from following key element：
Brand key element, generic name of the goods key element, unit key element, attribute key element and model key element.
8. keyword put-on method as claimed in claim 7, it is characterised in that step S14In comprise the following steps：
Step S141, filter out the key element keyword of the unit key element, the attribute key element and the model key element；
Step S142, calculate through step S141Investment yield of remaining each key element keyword after filtering out, and select to invest
Earning rate reaches the key element keyword of the Second Threshold as the keyword.
9. the keyword put-on method as described in claim l, it is characterised in that step S1Comprise the following steps：
Step S11, from a search history database obtain search key, have recorded user in the search history database searches
The search rate of the search key of rope and each search key；
Step S12, filter out a part in the search key, the part is：Preposition, auxiliary word, unit, attribute and model；
Step S13, will be through step S12Search key after filtering out is as the keyword.
10. keyword put-on method as claimed in claim 9, it is characterised in that step S3Comprise the following steps：
Step S31, judge whether the transition probability of each keyword reaches the first threshold, if so, into step S32；If it is not, enter
Enter step S33；
Step S32, the keyword launched by SEM channels；
Step S33, judge whether the search rate of the keyword reaches one the 3rd threshold value, if so, into step S34；If it is not, into
Step S35；
Step S34, the keyword launched by SEO channels；
Step S35, abandon the dispensing of the keyword.
11. a kind of keyword jettison system, it is characterised in that it includes：
One keyword acquisition device, for obtaining keyword to be put；
One computing device, for calculating the transition probability of the keyword；Transition probability represents that user is searched for after the keyword, purchases
The probability of the commodity related to the keyword is bought, the calculation formula of transition probability is：(purchase number/search results pages are clicked on secondary
Number) * 100%；
One keyword delivery device, the keyword for transition probability to be reached to a first threshold are launched by SEM channels.
12. keyword jettison system as claimed in claim 11, it is characterised in that the keyword delivery device is additionally operable to turn
The keyword that rate is not up to the first threshold is launched by SEO channels.
13. keyword jettison system as claimed in claim 11, it is characterised in that the keyword acquisition device includes：
One information acquisition device, determine to produce for obtaining product information from a product database, and according to the product information
Category and according to the product category does not call dictionary corresponding with the product category；
One key element generating means, for the product information to be divided into at least one key element；
One cutting word device, for using the dictionary and according to each key element that product information cutting word is crucial at least one key element
Word, and at least one key element keyword is formatted with the form of unified key element keyword；
One keyword generating means, for calculating an investment yield of each key element keyword, and investment yield is selected to reach
To a Second Threshold key element keyword as the keyword.
14. keyword jettison system as claimed in claim 13, it is characterised in that the cutting word device is additionally operable to using together
Adopted word dictionary unifies the mode of synonym, one by the way of unified capital and small letter and in replacing by the way of half double byte character
Kind or a variety of format at least one key element keyword with the form of unified key element keyword.
15. keyword jettison system as claimed in claim 13, it is characterised in that the dictionary is brand dictionary, commodity are general
One or more in title dictionary, unit dictionary, attribute dictionary, model dictionary and conventional word dictionary.
16. keyword jettison system as claimed in claim 15, it is characterised in that the cutting word device is additionally operable to use conventional word
Preposition and/or auxiliary word in dictionary excision product information.
17. keyword jettison system as claimed in claim 13, it is characterised in that at least one key element is selected from following want
Element：Brand key element, generic name of the goods key element, unit key element, attribute key element and model key element.
18. keyword jettison system as claimed in claim 17, it is characterised in that the keyword generating means also include：
One filters out unit, for filtering out the key element keyword of the unit key element, the attribute key element and the model key element；
One computing unit, for calculating the investment yield of remaining each key element keyword after filtering out, and select to invest
Earning rate reaches the key element keyword of a Second Threshold as the keyword.
19. keyword jettison system as claimed in claim 11, it is characterised in that the keyword acquisition device includes：
One search information acquisition device, for obtaining search key from a search history database, the search history data
The search key of user's search and the search rate of each search key are have recorded in storehouse；
One filtering device, for filtering out the part in the search key, the part is：Preposition, auxiliary word, unit, attribute and
Model；
One keyword generating means, for using the search key after the filtering device filters out as the keyword.
20. keyword jettison system as claimed in claim 19, it is characterised in that the keyword delivery device includes：
One first judgment means, for judging whether the transition probability of each keyword reaches the first threshold, if so, enabling one
First delivery device；If it is not, enable one second judgment means；
First delivery device, for the keyword to be launched by SEM channels；
Whether second judgment means, the search rate for judging the keyword reach one the 3rd threshold value, if so, enabling one
Two delivery devices；If it is not, abandon the dispensing of the keyword；
Second delivery device, for the keyword to be launched by SEO channels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210129565.0A CN103377249B (en) | 2012-04-27 | 2012-04-27 | Keyword put-on method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210129565.0A CN103377249B (en) | 2012-04-27 | 2012-04-27 | Keyword put-on method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103377249A CN103377249A (en) | 2013-10-30 |
| CN103377249B true CN103377249B (en) | 2017-12-22 |
Family
ID=49462375
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210129565.0A Active CN103377249B (en) | 2012-04-27 | 2012-04-27 | Keyword put-on method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103377249B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104700295A (en) * | 2013-12-04 | 2015-06-10 | 航天信息股份有限公司 | Invoicing system with sales analysis function |
| CN104298789B (en) * | 2014-11-13 | 2018-04-03 | 北京国双科技有限公司 | The division methods and device of keyword |
| CN104462412A (en) * | 2014-12-11 | 2015-03-25 | 北京国双科技有限公司 | Keyword detection method and device for release of internet keywords |
| CN106294416B (en) * | 2015-05-25 | 2019-10-15 | 阿里巴巴集团控股有限公司 | Disaggregated model method for building up, keyword choosing method and the device of SEO dictionary |
| CN106682039B (en) * | 2015-11-10 | 2020-10-27 | 北京国双科技有限公司 | Method and device for monitoring keyword putting effect |
| CN106776665A (en) * | 2015-11-25 | 2017-05-31 | 北京国双科技有限公司 | The put-on method and device of keyword |
| CN105956013A (en) * | 2016-04-21 | 2016-09-21 | 世纪禾光科技发展（北京）有限公司 | Method, device, and system for extracting website keyword |
| CN106033583A (en) * | 2016-05-18 | 2016-10-19 | 杭州算子科技有限公司 | Advertisement paid listing method and system applied to electronic commerce |
| CN106161627B (en) * | 2016-07-12 | 2019-10-15 | 百度在线网络技术（北京）有限公司 | Method and device for pushed information |
| CN108804430A (en) * | 2017-04-26 | 2018-11-13 | 广东原昇信息科技有限公司 | A kind of SEM launches data sorting system and its sorting technique |
| CN109993551A (en) * | 2017-12-29 | 2019-07-09 | 北京奇虎科技有限公司 | Determine that the method, apparatus and computer readable storage medium of crowd are launched in advertisement |
| CN108960917B (en) * | 2018-06-29 | 2022-06-03 | 深圳市跨境邦电子商务有限公司 | Network marketing promotion method and device |
| CN110544110A (en) * | 2019-07-30 | 2019-12-06 | 中至数据集团股份有限公司 | SEM keyword release abnormity monitoring method and device and readable storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129431A (en) * | 2010-01-13 | 2011-07-20 | 阿里巴巴集团控股有限公司 | Search method and system applied to online trading platform |
| CN102360484A (en) * | 2011-09-30 | 2012-02-22 | 奇智软件（北京）有限公司 | Method and apparatus for authenticity detection on sale data of group purchasing website |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8527352B2 (en) * | 2006-10-30 | 2013-09-03 | Adchemy, Inc. | System and method for generating optimized bids for advertisement keywords |
| US20110040616A1 (en) * | 2009-08-14 | 2011-02-17 | Yahoo! Inc. | Sponsored search bid adjustment based on predicted conversion rates |
| US8266006B2 (en) * | 2009-11-03 | 2012-09-11 | Ebay Inc. | Method, medium, and system for keyword bidding in a market cooperative |
-
2012
- 2012-04-27 CN CN201210129565.0A patent/CN103377249B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129431A (en) * | 2010-01-13 | 2011-07-20 | 阿里巴巴集团控股有限公司 | Search method and system applied to online trading platform |
| CN102360484A (en) * | 2011-09-30 | 2012-02-22 | 奇智软件（北京）有限公司 | Method and apparatus for authenticity detection on sale data of group purchasing website |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103377249A (en) | 2013-10-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103377249B (en) | Keyword put-on method and system | |
| KR102075833B1 (en) | Curation method and system for recommending of art contents | |
| CN110633373B (en) | Automobile public opinion analysis method based on knowledge graph and deep learning | |
| CN105630768B (en) | A kind of product name recognition method and device based on stacking condition random field | |
| CN102375823B (en) | Searching result gathering display method and system | |
| TWI631474B (en) | Method and device for product identification label and method for product navigation | |
| CN102609523B (en) | The collaborative filtering recommending method classified based on taxonomy of goods and user | |
| US20140195348A1 (en) | Method and apparatus for composing search phrases, distributing ads and searching product information | |
| CN107609052A (en) | A kind of generation method and device of the domain knowledge collection of illustrative plates based on semantic triangle | |
| US20130046771A1 (en) | Systems and methods for facilitating the gathering of open source intelligence | |
| CN105718184A (en) | Data processing method and apparatus | |
| US20140258002A1 (en) | Semantic model based targeted search advertising | |
| CN105868255A (en) | Query recommendation method and apparatus | |
| CN107665217A (en) | A kind of vocabulary processing method and system for searching service | |
| CN104391883B (en) | A kind of online advertisement audient's sort method based on transfer learning | |
| CN102254043A (en) | Semantic mapping-based clothing image retrieving method | |
| CN109960756A (en) | Media event information inductive method | |
| CN103942347A (en) | Word separating method based on multi-dimensional comprehensive lexicon | |
| CN105468649B (en) | Method and device for judging matching of objects to be displayed | |
| CN107958406A (en) | Inquire about acquisition methods, device and the terminal of data | |
| CN112258061B (en) | Intelligent risk analysis early warning system and early warning method for whole process of project | |
| CN106326318A (en) | Search method and device | |
| CN106682963A (en) | Recommendation system data completion method based on convex optimization local low-rank matrix approximation | |
| CN106897437A (en) | The many sorting techniques of high-order rule and its system of a kind of knowledge system | |
| CN110442670B (en) | Consumer portrait generation method based on text indexing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20160914 Address after: East Building 11, 100195 Beijing city Haidian District xingshikou Road No. 65 west Shan creative garden district 1-4 four layer of 1-4 layer Applicant after: Beijing Jingdong Shangke Information Technology Co., Ltd. Address before: 201203 Shanghai city Pudong New Area Zu Road No. 295 Room 102 Applicant before: Niuhai Information Technology (Shanghai) Co., Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |