CN108776901A - Method and system for advertisement recommendation based on search term - Google Patents
Method and system for advertisement recommendation based on search term Download PDFInfo
- Publication number
- CN108776901A CN108776901A CN201810392779.4A CN201810392779A CN108776901A CN 108776901 A CN108776901 A CN 108776901A CN 201810392779 A CN201810392779 A CN 201810392779A CN 108776901 A CN108776901 A CN 108776901A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- recommended
- word
- current search
- 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.)
- Granted
Links
Classifications
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to advertisement putting fields, and in particular to the method and system for advertisement recommendation based on search term.The method includes：It is extended processing for current search word input by user, obtains the extension phrase of current search word；Expansion processing is carried out for the text key word of every advertisement to be recommended, obtains the expansion phrase of every advertisement to be recommended；According to the expansion phrase of the extension phrase and every advertisement to be recommended of current search word, the comprehensive score that every advertisement to be recommended is directed to current search word is calculated；It is directed to the comprehensive score of current search word according to every advertisement to be recommended, the corresponding page that at least one recommended advertisements are illustrated in current search word is selected from advertisement to be recommended.The present invention can recommend the advertisement strong with search term degree of correlation, accomplish accurately to launch.
Description
Technical field
The present invention relates to advertisement putting fields, and in particular to the method and system for advertisement recommendation based on search term.
Background technology
In the prior art, screening advertisement is matched by search term itself；That is, according to search term with advertisement text
Matching degree determine correlation, further according to the height of correlation advertisement is ranked up, it is final to obtain the advertisement recommended.
The prior art has following problem：
1, search term is usually very short, and extension is big, and information is few, can cause to recall advertisement few.
2, word matching degree cannot reflect degree of correlation of the advertisement with search；For example search term is " internet conference ", is contained
Have " interconnection "；A certain advertisement is the product of a router, and product description also has " interconnection ".The degree of correlation of the two is in fact very
It is low.But it is matched, is inputted " internet conference " according to word, the advertisement for recommending the router will be corresponded to.In this way, may result in recommendation
Advertisement and search term degree of correlation itself it is not strong, influence the accurate dispensing of advertisement.
3, word unmatches, and flow is possible to be wasted.
Invention content
The technical problem to be solved in the present invention is, overcomes the shortcomings of existing technology, provides the advertisement based on search term
Recommend method and system, the advertisement strong with search term degree of correlation can be recommended, accomplish accurately to launch.
To reach above-mentioned technical purpose, on the one hand, method is recommended in the advertisement provided by the invention based on search term, including：
It is extended processing for current search word input by user, obtains the extension phrase of current search word；
Expansion processing is carried out for the text key word of every advertisement to be recommended, obtains the expansion word of every advertisement to be recommended
Group；
According to current search word extension phrase and every advertisement to be recommended expansion phrase, be calculated every it is to be recommended
Advertisement is directed to the comprehensive score of current search word；
It is directed to the comprehensive score of current search word according to every advertisement to be recommended, at least one is selected from advertisement to be recommended
Recommended advertisements are illustrated in the corresponding page of current search word.
On the other hand, the advertisement commending system provided by the invention based on search term, including：
Expanding element is extended processing for being directed to current search word input by user, obtains the expansion of current search word
Open up phrase；
Expansion unit carries out expansion processing for the text key word for every advertisement to be recommended, obtains every and wait pushing away
Recommend the expansion phrase of advertisement；
Computing unit is calculated for the expansion phrase for extending phrase and every advertisement to be recommended according to current search word
Obtain the comprehensive score that every advertisement to be recommended is directed to current search word；
Recommendation unit, the comprehensive score for being directed to current search word according to every advertisement to be recommended, from advertisement to be recommended
In select the corresponding page that at least one recommended advertisements are illustrated in current search word.
In the present invention, search term and advertisement to be recommended are extended and are expanded respectively, be expanded phrase and expansion
Phrase；Again by extending phrase and expanding the comprehensive score of phrase calculating search term and advertisement to be recommended；To by comprehensive
Divide to choose recommended advertisements.The present invention can match more advertisements after extending search term as a result, so that recommend wide
Announcement has more more options；So that having stronger relevance between recommended advertisements and search term, so as to accomplish the accurate throwing of advertisement
It puts.
Description of the drawings
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 technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the method flow schematic diagram of the embodiment of the present invention；
Fig. 2 is the system structure diagram of the embodiment of the present invention；
Fig. 3 is specific works flow process schematic diagram in the embodiment of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Embodiment 1.1：As shown in Figure 1, method is recommended in the advertisement of the present invention based on search term, including：
101, it is extended processing for current search word input by user, obtains the extension phrase of current search word；
102, expansion processing is carried out for the text key word of every advertisement to be recommended, obtains the expansion of every advertisement to be recommended
Fill phrase；
103, it according to the expansion phrase of the extension phrase and every advertisement to be recommended of current search word, is calculated every and waits for
Recommended advertisements are directed to the comprehensive score of current search word；
104, it is directed to the comprehensive score of current search word according to every advertisement to be recommended, is selected at least from advertisement to be recommended
One recommended advertisements is illustrated in the corresponding page of current search word.
Embodiment 1.2：Embodiment 1.1 is connect, it is described to be extended processing for current search word input by user, worked as
The extension phrase of preceding search term further includes before：
Judge that current search word input by user searches word for heat；It is the search that its word frequency is more than predetermined word frequency that the heat, which searches word,
Word.
That is, when it is that heat searches word to judge current search word input by user, executes and work as input by user
Preceding search term is extended processing, the step of obtaining the extension phrase of current search word.
Embodiment 1.3：Embodiment 1.1 or 1.2 is connect, it is described to be extended processing for current search word input by user,
The extension phrase of current search word is obtained, further includes later：
The extension phrase of current search word is segmented successively, individual character is removed and stop words is gone to handle；
The text key word for every advertisement to be recommended carries out expansion processing, obtains the expansion of every advertisement to be recommended
Phrase is filled, further includes later：
Expansion phrase corresponding to the text key word of every advertisement to be recommended is segmented, removes individual character and go to deactivate successively
Word processing.
In the above-mentioned technical solutions, described to be extended processing for current search word input by user, currently searched
The extension phrase of rope word, specifically includes：
The expansion word of current search word is obtained, the expansion word includes but not limited to：Related term, synonym, is total to impression word
Existing word, conjunctive word and extended log files word；
Current search word is combined into the extension phrase of current search word with corresponding expansion word.
The related term, the word inputted in predetermined amount of time after inputting current search word for same user；
The impression word is the word for having public sentiment relationship with current search word；
The co-occurrence word, to input the word in search box simultaneously with current search word；
The conjunctive word, to have the word of incidence relation with current search word in word sequence；
The extended log files word, to check the context of current search word by searching for daily record, within a context word frequency compared with
High word.
In the above-mentioned technical solutions, the expansion word of the extension phrase and every advertisement to be recommended according to current search word
Group is calculated the comprehensive score that every advertisement to be recommended is directed to current search word, specifically includes：
The expansion phrase of the extension phrase of current search word and every advertisement to be recommended is encoded respectively, is obtained current
The vector of the advertisement to be recommended of vector sum every of search term；
Vectorial cosine similarity of the vector for current search word for calculating separately every advertisement to be recommended, to obtain
Every advertisement to be recommended is directed to the similar score of word of current search word；
Obtain each attribute tags of the extension phrase of current search word；
Judge each attribute tags of the expansion phrase of every advertisement to be recommended；
Each attribute tags of the extension phrase of current search word are each with the expansion phrase of every advertisement to be recommended respectively
Attribute tags are matched, and the similar score of attribute that every advertisement to be recommended is directed to current search word is calculated；
The similar score of word score similar with attribute by described every advertisement to be recommended for current search word is added
Or weighting summation, obtain the comprehensive score that every advertisement to be recommended is directed to current search word.
Further, the attribute tags include：Theme label, type label and industry label；
Each attribute tags of the extension phrase for obtaining current search word, specifically include：
According to historical data, agent model, Type model and business models are established respectively；
The expansion word group of current search word is inputted into topic model, Type model and business models respectively, output obtains pair
Answer theme label, type label and industry label；
Each attribute tags expansion phrase with every advertisement to be recommended respectively of the extension phrase by current search word
Each attribute tags matched, be calculated every advertisement to be recommended be directed to current search word the similar score of attribute, specifically
Including：
Compare the theme mark of the theme label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word
Label；
If the theme label of the theme label of the extension phrase of current search word and the expansion phrase of current advertisement to be recommended
It is identical, then the first predetermined score is included in the theme score that current advertisement to be recommended is directed to current search word；
Compare the type mark of the type label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word
Label；
If the theme label of the theme label of the extension phrase of current search word and the expansion phrase of current advertisement to be recommended
It is identical, then the second predetermined score is included in the type scores that current advertisement to be recommended is directed to current search word；
Compare the industry mark of the industry label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word
Label；
If the industry label of the industry label of the extension phrase of current search word and the expansion phrase of current advertisement to be recommended
It is identical, then third predetermined score is included in the industry score that current advertisement to be recommended is directed to current search word；
Current advertisement to be recommended is added or is weighted with industry score for the theme score of current search word, type scores
It is added, obtains the similar score of attribute that current advertisement to be recommended is directed to current search word.
In the above-mentioned technical solutions, the comprehensive score of current search word is directed to according to every advertisement to be recommended, to be recommended
The corresponding page that at least one recommended advertisements are illustrated in current search word is selected in advertisement, is specifically included：
Every advertisement to be recommended is directed to the comprehensive score of current search word by it, is ranked up from high to low；
Take the advertisement to be recommended of preceding n (n > 0) items as recommended advertisements；
Recommended advertisements are illustrated in the corresponding page of current search word according to described sort.
Embodiment 2.1：As shown in Fig. 2, the advertisement commending system of the present invention based on search term, including：
Expanding element 11 is extended processing for being directed to current search word input by user, obtains current search word
Extend phrase；
Expansion unit 12 carries out expansion processing for the text key word for every advertisement to be recommended, obtains every and wait for
The expansion phrase of recommended advertisements；
Computing unit 13, for the expansion phrase for extending phrase and every advertisement to be recommended according to current search word, meter
It calculates and obtains the comprehensive score that every advertisement to be recommended is directed to current search word；
Recommendation unit 14, the comprehensive score for being directed to current search word according to every advertisement to be recommended, to be recommended wide
The corresponding page that at least one recommended advertisements are illustrated in current search word is selected in announcement.
Embodiment 2.2：Embodiment 2.1 is connect, as a kind of possible structure, the system also includes：
Judging unit 15 triggers expanding element 11 when for judging that current search word input by user searches word for heat；It is described
It is the search term that its word frequency is more than predetermined word frequency that heat, which searches word,.
Embodiment 2.3：Embodiment 2.1 or 2.2 is connect, as alternatively possible structure, the system also includes：
First filter element 16 is segmented for the extension phrase to current search word, removes individual character and go to deactivate successively
Word processing；
Second filter element 17 carries out successively for the corresponding expansion phrase of the text key word to every advertisement to be recommended
It segments, remove individual character and stop words is gone to handle.
In the above-mentioned technical solutions, the expanding element 11, is specifically used for：
The expansion word of current search word is obtained, the expansion word includes but not limited to：Related term, synonym, is total to impression word
Existing word, conjunctive word and extended log files word；
Current search word is combined into the extension phrase of current search word with corresponding expansion word.
In the above-mentioned technical solutions, the computing unit 13, including：
Coding module, for being carried out respectively to the expansion phrase of the extension phrase of current search word and every advertisement to be recommended
Coding, obtains the vector of the advertisement to be recommended of vector sum every of current search word；
Computing module, vectorial cosine phase of the vector for current search word for calculating separately every advertisement to be recommended
Like degree, to obtain the similar score of word that every advertisement to be recommended is directed to current search word；
Acquisition module, each attribute tags of the extension phrase for obtaining current search word；
Determination module, each attribute tags of the expansion phrase for judging every advertisement to be recommended；
Matching module, for by each attribute tags of the extension phrase of current search word respectively with every advertisement to be recommended
Each attribute tags for expanding phrase are matched, and the attribute that every advertisement to be recommended is calculated for current search word is similar
Point；
Summation module, for described every advertisement to be recommended is similar with attribute for the similar score of word of current search word
Score carries out addition or weighting summation, obtains the comprehensive score that every advertisement to be recommended is directed to current search word.
Further, the attribute tags include：Theme label, type label and industry label；
The acquisition module, is specifically used for：
According to historical data, agent model, Type model and business models are established respectively；
The expansion word group of current search word is inputted into topic model, Type model and business models respectively, output obtains pair
Answer theme label, type label and industry label；
The matching module, is specifically used for：
Compare the theme mark of the theme label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word
Label；
If the theme label of the theme label of the extension phrase of current search word and the expansion phrase of current advertisement to be recommended
It is identical, then the first predetermined score is included in the theme score that current advertisement to be recommended is directed to current search word；
Compare the type mark of the type label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word
Label；
If the theme label of the theme label of the extension phrase of current search word and the expansion phrase of current advertisement to be recommended
It is identical, then the second predetermined score is included in the type scores that current advertisement to be recommended is directed to current search word；
Compare the industry mark of the industry label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word
Label；
If the industry label of the industry label of the extension phrase of current search word and the expansion phrase of current advertisement to be recommended
It is identical, then third predetermined score is included in the industry score that current advertisement to be recommended is directed to current search word；
Current advertisement to be recommended is added or is weighted with industry score for the theme score of current search word, type scores
It is added, obtains the similar score of attribute that current advertisement to be recommended is directed to current search word.
In the above-mentioned technical solutions, the recommendation unit 14, is specifically used for：
Every advertisement to be recommended is directed to the comprehensive score of current search word by it, is ranked up from high to low；
Take the advertisement to be recommended of preceding n (n > 0) items as recommended advertisements；
Recommended advertisements are illustrated in the corresponding page of current search word according to described sort.
Above-mentioned technical proposal of the embodiment of the present invention is described in detail below in conjunction with application example：
By taking microblogging as an example；
As shown in figure 3, method is recommended in the advertisement based on search term, including：
Step 1, judgement current search word input by user search word for heat；It is more than predetermined word frequency that the heat, which searches word as its word frequency,
Search term；
It is huge that user enters the volumes of searches of microblogging search box daily, if each search term is extended matching extensively
If announcement, the burden of server will weigh very much.The hot topic for showing to occur with the proximal segment time in microblogging search box by statistics
Time, related search term accounted for the overwhelming majority, so the frequency of occurrences of certain search terms can be very high, this word is referred to as heat and searches
Word.Therefore, in this example, extension and matching advertisement after word carries out only is searched to heat, can both substantially reduce micro blog server
Pressure, can also improve advertisement dispensing efficiency.
Step 2 judges whether current search word input by user and current advertisement to be recommended are stored with comprehensive score；
In real work realization, all each search terms input by user are required for and all advertisement meters to be recommended
Calculate comprehensive score.The workload calculated in this way is huge, since different user is directed to some hot ticket in certain time period
Identical search term is inputted, in fact, the search term, which is also referred to as heat, searches word, so in above-mentioned evaluation work, is had a large amount of
It computes repeatedly.
Therefore, the comprehensive score between same search term and same advertisement to be recommended is stored to comprehensive score table, is being counted
Before the comprehensive score for calculating current search word and current advertisement to be recommended, first inquire in comprehensive score table whether have current search word
With the comprehensive score of current advertisement to be recommended.
If current search word and current advertisement to be recommended are stored with comprehensive score, the comprehensive score is directly returned；
If current search word and current advertisement to be recommended are not stored with comprehensive score, also indicate that, did not calculate comprehensive
Score then enters step 3.
It should be noted that in work is realized, the technique effect of the present invention can also be reached by omitting step 2.
Step 3 judges whether current search word input by user is stored with extension phrase；
Similarly step 2 is required for being extended in real work realization for each each search term input by user
Processing.This creates the terminal a large amount of workloads, so being extended that treated, search term (heat searches word) can all be stored to expansion
Open up phrase storage table.
If being stored with the extension phrase of current search word in extension phrase storage table, the extension phrase is returned directly to；
If there is no the extension phrase of current search word in extension phrase storage table, 4 are entered step.
It should be noted that in work is realized, the technique effect of the present invention can also be reached by omitting step 3.
Step 4 is extended processing for current search word input by user, obtains the extension phrase of current search word；
Specifically：
The expansion word of current search word is obtained, the expansion word includes but not limited to：Related term, synonym, is total to impression word
Existing word, conjunctive word and extended log files word；
Current search word is combined into the extension phrase of current search word with corresponding expansion word.
The related term, the word inputted in predetermined amount of time after inputting current search word for same user；
The impression word is the word for having public sentiment relationship with current search word；
The co-occurrence word, to input the word in search box simultaneously with current search word；
The conjunctive word, to have the word of incidence relation with current search word in word sequence；
The extended log files word, to check the context of current search word by searching for daily record, within a context word frequency compared with
High word.
In real work realization, the expansion word of current search word can not also be limited to above-mentioned word.
Step 5 judges whether current advertisement to be recommended is stored with expansion phrase；
Similarly step 3 is required for carrying out expansion processing in real work realization for each advertisement to be recommended.In this way
A large amount of workload is just produced, treated that advertisement to be recommended can all store to expanding phrase storage table so expand.
If expanding the extension phrase for being stored with current advertisement to be recommended in phrase storage table, the expansion word is returned directly to
Group；
If expanding does not have the extension phrase of current advertisement to be recommended in phrase storage table, 6 are entered step.
It should be noted that in work is realized, the technique effect of the present invention can also be reached by omitting step 5.
Step 6 carries out expansion processing for the text key word of every advertisement to be recommended, obtains every advertisement to be recommended
Expand phrase；
Before treating recommended advertisements and carrying out expansion processing, need to find the keyword in current advertisement to be recommended, it is then right
The keyword carries out expansion processing, obtains the expansion phrase of current advertisement to be recommended.
Step 7 segments the extension phrase of current search word, removes individual character and stop words is gone to handle successively；
Step 8, expansion phrase corresponding to the text key word of every advertisement to be recommended segmented successively, go individual character and
Stop words is gone to handle；
When being segmented to extension phrase and expansion phrase, participle profession may be used and turn part.Going individual character to be exactly will extension
Phrase and the word for expanding the independent word formation occurred in phrase remove.Stop words, generally by developer according to practical work
Make situation determination, in general, including：,, a, b, c etc. be free of the word of specific meaning.
Step 9, the expansion phrase for extending phrase and every advertisement to be recommended according to current search word, are calculated every
Advertisement to be recommended is directed to the comprehensive score of current search word；Specifically：
Step 9.1 respectively encodes the expansion phrase of the extension phrase of current search word and every advertisement to be recommended,
Obtain the vector of the advertisement to be recommended of vector sum every of current search word；
For example, the extension phrase of current search word is：" certain star, son, birth, milk father "；
The expansion phrase of current advertisement to be recommended is：" Germany, milk powder, milk father "；
Current extensions phrase and expansion phrase are encoded；
First, merge current extensions phrase and expand phrase, obtain combination phrase：" certain star, son, birth, milk father,
Germany, milk powder "；
Then, current extensions phrase is compared with combination phrase, encounters identical word and be encoded to 1, encounters difference
Word be encoded to 0；The vectorial A for obtaining current search word is【1,1,1,1,0,0】；
The current phrase that expands is compared with combination phrase, identical word is encountered and is encoded to 1, encounter different words
It is encoded to 0；The vectorial B for obtaining current search word is【0,0,0,1,1,1】；
Step 9.2, vectorial cosine similarity of the vector for current search word for calculating separately every advertisement to be recommended,
To obtain the similar score of word that every advertisement to be recommended is directed to current search word；
With Euclidean distance, the cosine similarity of A and B is calculated；Such as following formula：
Similarity=cos (theta)=AB/ | | A | | | | B | | (1)
Step 9.3, each attribute tags for extending phrase for obtaining current search word；
According to historical data, agent model, Type model and business models are established respectively；
The expansion word group of current search word is inputted into topic model, Type model and business models respectively, output obtains pair
Answer theme label, type label and industry label.
In theme level, the heat for combining history searches the text inner cylinder of data and advertisement, and model LDA etc. is generated with document subject matter
Training topic model.Using the extension phrase of current search word as the input of topic model, output obtains at least one theme.
Predefine out various types, disaggregated model established according to historical data, typically, with support vector machines svm or
BayesBayes classifies.
Step 9.4, each attribute tags for expanding phrase of every advertisement to be recommended of judgement；
Step 9.5, each attribute tags expansion with every advertisement to be recommended respectively for extending phrase by current search word
Each attribute tags of phrase are matched, and the similar score of attribute that every advertisement to be recommended is directed to current search word is calculated；
Specifically：
Compare the theme mark of the theme label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word
Label；
If the theme label of the theme label of the extension phrase of current search word and the expansion phrase of current advertisement to be recommended
It is identical, then the first predetermined score is included in the theme score that current advertisement to be recommended is directed to current search word；
Compare the type mark of the type label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word
Label；
If the theme label of the theme label of the extension phrase of current search word and the expansion phrase of current advertisement to be recommended
It is identical, then the second predetermined score is included in the type scores that current advertisement to be recommended is directed to current search word；
Compare the industry mark of the industry label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word
Label；
If the industry label of the industry label of the extension phrase of current search word and the expansion phrase of current advertisement to be recommended
It is identical, then third predetermined score is included in the industry score that current advertisement to be recommended is directed to current search word；
Current advertisement to be recommended is added or is weighted with industry score for the theme score of current search word, type scores
It is added, obtains the similar score of attribute that current advertisement to be recommended is directed to current search word.
It is compared after topic model exports to obtain at least one theme, and with the theme of current advertisement to be recommended.If having one
A theme matching, if the first predetermined score is 10 points, then current advertisement to be recommended is scored at 10 for the theme of current search word
Point.If there are two themes to match, current advertisement to be recommended is scored at 20 points for the theme of current search word；And so on.
The calculating similarly calculating with theme score for type scores.Type is all predefined, and expansion word is searched to heat
The classification for the blog article for directly branching away classification followed by advertisement with disaggregated model is compared, if it does, then predetermined plus second
Score.If using hierarchical classification system, the second predetermined score can be taken and added by the Relevance scores between different layers
Weight average is worth to.
Step 9.6, by the similar score of institute's predicate score similar with each attribute, be added or weighting summation obtain it is comprehensive
Point.
Step 10, the comprehensive score that current search word is directed to according to every advertisement to be recommended, are selected from advertisement to be recommended
At least one recommended advertisements are illustrated in the corresponding page of current search word；Specifically：
Every advertisement to be recommended is directed to the comprehensive score of current search word by step 10.1 by it, is arranged from high to low
Sequence；
Step 10.2 takes the advertisement to be recommended of preceding n (n > 0) items as recommended advertisements；
It can also select composite score more than each advertisement to be recommended of some predetermined score as recommended advertisements here.
Recommended advertisements are illustrated in the corresponding page of current search word by step 10.3 according to described sort.
In the prior art, matching search term and when advertisement to be recommended, the only limitation of keyword, multiple keywords it
Between without sequence, lead to the semantic meaning representation for lacking Chinese language model n-gram or sentence level.In the present invention.Master is added
After the information such as topic, classification and industry, more semantic matches can be generated, it is more matched to be found by current search word
Advertisement makes advertisement accurately be launched, and the increase for making advertisement recall.
It should be understood that the particular order or level of the step of during disclosed are the examples of illustrative methods.Based on setting
Count preference, it should be appreciated that in the process the step of particular order or level can be in the feelings for the protection domain for not departing from the disclosure
It is rearranged under condition.Appended claim to a method is not illustratively sequentially to give the element of various steps, and not
It is to be limited to the particular order or level.
In above-mentioned detailed description, various features are combined together in single embodiment, to simplify the disclosure.No
This published method should be construed to reflect such intention, that is, the embodiment of theme claimed needs to compare
The more features of feature clearly stated in each claim.On the contrary, as appended claims is reflected
Like that, the present invention is in the state fewer than whole features of disclosed single embodiment.Therefore, appended claims
It is hereby expressly incorporated into detailed description, wherein each claim is used as alone the individual preferred embodiment of the present invention.
For so that any technical staff in the art is realized or using the present invention, above to disclosed embodiment into
Description is gone.To those skilled in the art；The various modifications mode of these embodiments will be apparent from, and this
The General Principle of text definition can also be suitable for other embodiments on the basis of not departing from the spirit and scope of the disclosure.
Therefore, the disclosure is not limited to embodiments set forth herein, but most wide with principle disclosed in the present application and novel features
Range is consistent.
Described above includes the citing of one or more embodiments.Certainly, in order to describe above-described embodiment and description portion
The all possible combination of part or method is impossible, but it will be appreciated by one of ordinary skill in the art that each implementation
Example can do further combinations and permutations.Therefore, embodiment described herein is intended to cover fall into the appended claims
Protection domain in all such changes, modifications and variations.In addition, with regard to the term used in specification or claims
The mode that covers of "comprising", the word is similar to term " comprising ", just as " including " solved in the claims as link word
As releasing.In addition, the use of any one of specification in claims term "or" being to indicate " non-exclusionism
Or ".
Those skilled in the art will also be appreciated that the various illustrative components, blocks that the embodiment of the present invention is listed
(illustrative logical block), unit and step can pass through the knot of electronic hardware, computer software, or both
Conjunction is realized.To clearly show that the replaceability (interchangeability) of hardware and software, above-mentioned various explanations
Property component (illustrative components), unit and step universally describe their function.Such work(
Can be that the design requirement for depending on specific application and whole system is realized by hardware or software.Those skilled in the art
Can be for each specific function of applying, the realization of various methods can be used described, but this realization is understood not to
Range beyond protection of the embodiment of the present invention.
Various illustrative logical blocks or unit described in the embodiment of the present invention can by general processor,
Digital signal processor, application-specific integrated circuit (ASIC), field programmable gate array or other programmable logic devices, discrete gate
Or described function is realized or is operated in transistor logic, the design of discrete hardware components or any of the above described combination.General place
It can be microprocessor to manage device, and optionally, which may be any traditional processor, controller, microcontroller
Device or state machine.Processor can also be realized by the combination of computing device, such as digital signal processor and microprocessor,
Multi-microprocessor, one or more microprocessors combine a digital signal processor core or any other like configuration
To realize.
The step of method described in the embodiment of the present invention or algorithm can be directly embedded into hardware, processor execute it is soft
The combination of part module or the two.Software module can be stored in RAM memory, flash memory, ROM memory, EPROM storages
Other any form of storaging mediums in device, eeprom memory, register, hard disk, moveable magnetic disc, CD-ROM or this field
In.Illustratively, storaging medium can be connect with processor, so that processor can read information from storaging medium, and
It can be to storaging medium stored and written information.Optionally, storaging medium can also be integrated into processor.Processor and storaging medium can
To be set in ASIC, ASIC can be set in user terminal.Optionally, processor and storaging medium can also be set to use
In different components in the terminal of family.
In one or more illustrative designs, above-mentioned function described in the embodiment of the present invention can be in hardware, soft
Part, firmware or the arbitrary of this three combine to realize.If realized in software, these functions can store and computer-readable
On medium, or with one or more instruction or code form be transmitted on the medium of computer-readable.Computer readable medium includes electricity
Brain storaging medium and convenient for allow computer program to be transferred to from a place telecommunication media in other places.Storaging medium can be with
It is that any general or special computer can be with the useable medium of access.For example, such computer readable media may include but
It is not limited to RAM, ROM, EEPROM, CD-ROM or other optical disc storage, disk storage or other magnetic storage devices or other
What can be used for carry or store with instruct or data structure and it is other can be by general or special computer or general or specially treated
The medium of the program code of device reading form.In addition, any connection can be properly termed computer readable medium, example
Such as, if software is to pass through a coaxial cable, fiber optic cables, double from a web-site, server or other remote resources
Twisted wire, Digital Subscriber Line (DSL) are defined with being also contained in for the wireless way for transmitting such as example infrared, wireless and microwave
In computer readable medium.The disk (disk) and disk (disc) includes compress disk, radium-shine disk, CD, DVD, floppy disk
And Blu-ray Disc, disk is usually with magnetic duplication data, and disk usually carries out optical reproduction data with laser.Combinations of the above
It can also be included in computer readable medium.
Above-described specific implementation mode has carried out further the purpose of the present invention, technical solution and advantageous effect
It is described in detail, it should be understood that the foregoing is merely the specific implementation mode of the present invention, is not intended to limit the present invention
Protection domain, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all include
Within protection scope of the present invention.
Claims (14)
1. method is recommended in a kind of advertisement based on search term, which is characterized in that the method includes：
It is extended processing for current search word input by user, obtains the extension phrase of current search word；
Expansion processing is carried out for the text key word of every advertisement to be recommended, obtains the expansion phrase of every advertisement to be recommended；
According to the expansion phrase of the extension phrase and every advertisement to be recommended of current search word, every advertisement to be recommended is calculated
For the comprehensive score of current search word；
It is directed to the comprehensive score of current search word according to every advertisement to be recommended, at least one recommendation is selected from advertisement to be recommended
Corresponding page of the advertising display in current search word.
2. method is recommended in the advertisement according to claim 1 based on search term, which is characterized in that described to be inputted for user
Current search word be extended processing, obtain the extension phrase of current search word, further include before：
Judge that current search word input by user searches word for heat；It is the search term that its word frequency is more than predetermined word frequency that the heat, which searches word,.
3. method is recommended in the advertisement according to claim 1 or 2 based on search term, which is characterized in that described to be directed to user
The current search word of input is extended processing, obtains the extension phrase of current search word, specifically includes：
The expansion word of current search word is obtained, the expansion word includes but not limited to：Related term, impression word, synonym, co-occurrence
Word, conjunctive word and extended log files word；
Current search word is combined into the extension phrase of current search word with corresponding expansion word.
4. method is recommended in the advertisement according to claim 1 or 2 based on search term, which is characterized in that described to be directed to user
The current search word of input is extended processing, obtains the extension phrase of current search word, further includes later：
The extension phrase of current search word is segmented successively, individual character is removed and stop words is gone to handle；
The text key word for every advertisement to be recommended carries out expansion processing, obtains the expansion word of every advertisement to be recommended
Group further includes later：
Expansion phrase corresponding to the text key word of every advertisement to be recommended is segmented, removes individual character and go at stop words successively
Reason.
5. method is recommended in the advertisement according to claim 1 or 2 based on search term, which is characterized in that the basis is current
The expansion phrase of the extension phrase and every advertisement to be recommended of search term, is calculated every advertisement to be recommended and is directed to current search
The comprehensive score of word, specifically includes：
The expansion phrase of the extension phrase of current search word and every advertisement to be recommended is encoded respectively, obtains current search
The vector of the advertisement to be recommended of vector sum every of word；
Vectorial cosine similarity of the vector for current search word for calculating separately every advertisement to be recommended, to obtain every
Advertisement to be recommended is directed to the similar score of word of current search word；
Obtain each attribute tags of the extension phrase of current search word；
Judge each attribute tags of the expansion phrase of every advertisement to be recommended；
By each attribute tags of the extension phrase of current search word each attribute with the expansion phrase of every advertisement to be recommended respectively
Label is matched, and the similar score of attribute that every advertisement to be recommended is directed to current search word is calculated；
Described every advertisement to be recommended is added or is added for the similar score of the word score similar with attribute of current search word
Power is added, and obtains the comprehensive score that every advertisement to be recommended is directed to current search word.
6. method is recommended in the advertisement according to claim 5 based on search term, which is characterized in that the attribute tags packet
It includes：Theme label, type label and industry label；
Each attribute tags of the extension phrase for obtaining current search word, specifically include：
According to historical data, agent model, Type model and business models are established respectively；
The expansion word group of current search word is inputted into topic model, Type model and business models respectively, output obtains corresponding master
Inscribe label, type label and industry label；
Each attribute tags of the extension phrase by current search word are each with the expansion phrase of every advertisement to be recommended respectively
Attribute tags are matched, and the similar score of attribute that every advertisement to be recommended is directed to current search word is calculated, specifically includes：
Compare the theme label of the theme label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word；
If the theme label of the extension phrase of current search word is identical as the current expansion theme label of phrase of advertisement to be recommended,
The first predetermined score is then included in the theme score that current advertisement to be recommended is directed to current search word；
Compare the type label of the type label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word；
If the theme label of the extension phrase of current search word is identical as the current expansion theme label of phrase of advertisement to be recommended,
The second predetermined score is then included in the type scores that current advertisement to be recommended is directed to current search word；
Compare the industry label of the industry label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word；
If the industry label of the extension phrase of current search word is identical as the current expansion industry label of phrase of advertisement to be recommended,
Third predetermined score is then included in the industry score that current advertisement to be recommended is directed to current search word；
Current advertisement to be recommended is added or is weighted phase with industry score for the theme score of current search word, type scores
Add, obtains the similar score of attribute that current advertisement to be recommended is directed to current search word.
7. method is recommended in the advertisement according to claim 1 or 2 based on search term, which is characterized in that wait pushing away according to every
The comprehensive score that advertisement is directed to current search word is recommended, at least one recommended advertisements are selected from advertisement to be recommended is illustrated in and currently search
The corresponding page of rope word, specifically includes：
Every advertisement to be recommended is directed to the comprehensive score of current search word by it, is ranked up from high to low；
Take the advertisement to be recommended of preceding n (n > 0) items as recommended advertisements；
Recommended advertisements are illustrated in the corresponding page of current search word according to described sort.
8. a kind of advertisement commending system based on search term, which is characterized in that the system comprises：
Expanding element is extended processing for being directed to current search word input by user, obtains the expansion word of current search word
Group；
Expansion unit carries out expansion processing for the text key word for every advertisement to be recommended, obtains every to be recommended wide
The expansion phrase of announcement；
Computing unit is calculated for the expansion phrase for extending phrase and every advertisement to be recommended according to current search word
Every advertisement to be recommended is directed to the comprehensive score of current search word；
Recommendation unit, the comprehensive score for being directed to current search word according to every advertisement to be recommended, is selected from advertisement to be recommended
Go out the corresponding page that at least one recommended advertisements are illustrated in current search word.
9. the advertisement commending system according to claim 8 based on search term, which is characterized in that the system also includes：
Judging unit triggers the expanding element when for judging that current search word input by user searches word for heat；The heat is searched
Word is the search term that its word frequency is more than predetermined word frequency.
10. the advertisement commending system based on search term according to claim 8 or claim 9, which is characterized in that the extension is single
Member is specifically used for：
The expansion word of current search word is obtained, the expansion word includes but not limited to：Related term, impression word, synonym, co-occurrence
Word, conjunctive word and extended log files word；
Current search word is combined into the extension phrase of current search word with corresponding expansion word.
11. the advertisement commending system based on search term according to claim 8 or claim 9, which is characterized in that the system is also wrapped
It includes：
First filter element is segmented for the extension phrase to current search word, removes individual character and stop words is gone to handle successively；
Second filter element, segmented successively for the corresponding expansion phrase of the text key word to every advertisement to be recommended,
It removes individual character and stop words is gone to handle.
12. the advertisement commending system based on search term according to claim 8 or claim 9, which is characterized in that the calculating is single
Member, including：
Coding module, for being compiled respectively to the expansion phrase of the extension phrase of current search word and every advertisement to be recommended
Code, obtains the vector of the advertisement to be recommended of vector sum every of current search word；
Computing module, the vector for calculating separately every advertisement to be recommended are similar for the cosine of the vector of current search word
Degree, to obtain the similar score of word that every advertisement to be recommended is directed to current search word；
Acquisition module, each attribute tags of the extension phrase for obtaining current search word；
Determination module, each attribute tags of the expansion phrase for judging every advertisement to be recommended；
Matching module is used for each attribute tags expansion with every advertisement to be recommended respectively of the extension phrase of current search word
Each attribute tags of phrase are matched, and the similar score of attribute that every advertisement to be recommended is directed to current search word is calculated；
Summation module, the similar score of the word score similar with attribute for described every advertisement to be recommended to be directed to current search word
Addition or weighting summation are carried out, the comprehensive score that every advertisement to be recommended is directed to current search word is obtained.
13. the advertisement commending system according to claim 12 based on search term, which is characterized in that the attribute tags packet
It includes：Theme label, type label and industry label；
The acquisition module, is specifically used for：
According to historical data, agent model, Type model and business models are established respectively；
The expansion word group of current search word is inputted into topic model, Type model and business models respectively, output obtains corresponding master
Inscribe label, type label and industry label；
The matching module, is specifically used for：
Compare the theme label of the theme label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word；
If the theme label of the extension phrase of current search word is identical as the current expansion theme label of phrase of advertisement to be recommended,
The first predetermined score is then included in the theme score that current advertisement to be recommended is directed to current search word；
Compare the type label of the type label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word；
If the theme label of the extension phrase of current search word is identical as the current expansion theme label of phrase of advertisement to be recommended,
The second predetermined score is then included in the type scores that current advertisement to be recommended is directed to current search word；
Compare the industry label of the industry label and the expansion phrase of every advertisement to be recommended of the extension phrase of current search word；
If the industry label of the extension phrase of current search word is identical as the current expansion industry label of phrase of advertisement to be recommended,
Third predetermined score is then included in the industry score that current advertisement to be recommended is directed to current search word；
Current advertisement to be recommended is added or is weighted phase with industry score for the theme score of current search word, type scores
Add, obtains the similar score of attribute that current advertisement to be recommended is directed to current search word.
14. the advertisement commending system based on search term according to claim 8 or claim 9, which is characterized in that the recommendation is single
Member is specifically used for：
Every advertisement to be recommended is directed to the comprehensive score of current search word by it, is ranked up from high to low；
Take the advertisement to be recommended of preceding n (n > 0) items as recommended advertisements；
Recommended advertisements are illustrated in the corresponding page of current search word according to described sort.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810392779.4A CN108776901B (en) | 2018-04-27 | 2018-04-27 | Advertisement recommendation method and system based on search terms |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810392779.4A CN108776901B (en) | 2018-04-27 | 2018-04-27 | Advertisement recommendation method and system based on search terms |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108776901A true CN108776901A (en) | 2018-11-09 |
| CN108776901B CN108776901B (en) | 2021-01-15 |
Family
ID=64026649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810392779.4A Active CN108776901B (en) | 2018-04-27 | 2018-04-27 | Advertisement recommendation method and system based on search terms |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108776901B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109885753A (en) * | 2019-01-16 | 2019-06-14 | 苏宁易购集团股份有限公司 | A kind of method and device for expanding commercial articles searching and recalling |
| CN110222147A (en) * | 2019-05-15 | 2019-09-10 | 北京百度网讯科技有限公司 | Label extending method, device, computer equipment and storage medium |
| CN110706021A (en) * | 2019-09-12 | 2020-01-17 | 微梦创科网络科技（中国）有限公司 | Advertisement putting method and system |
| CN110781204A (en) * | 2019-09-09 | 2020-02-11 | 腾讯大地通途（北京）科技有限公司 | Identification information determination method, device, equipment and storage medium of target object |
| CN111383069A (en) * | 2018-12-29 | 2020-07-07 | 美宅科技（北京）有限公司 | Furniture assembly recommendation method and device |
| CN112256822A (en) * | 2020-10-21 | 2021-01-22 | 平安科技（深圳）有限公司 | Text search method and device, computer equipment and storage medium |
| CN112559693A (en) * | 2020-12-25 | 2021-03-26 | 鼎易创展咨询（北京）有限公司 | Similar word searching method, device, equipment and medium for text words |
| CN113111216A (en) * | 2020-01-13 | 2021-07-13 | 百度在线网络技术（北京）有限公司 | Advertisement recommendation method, device, equipment and storage medium |
| CN113297489A (en) * | 2021-05-28 | 2021-08-24 | 平安科技（深圳）有限公司 | Rehabilitation aid recommendation method and device, computer equipment and storage medium |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102063468A (en) * | 2010-12-03 | 2011-05-18 | 百度在线网络技术（北京）有限公司 | Device and method for confirming query sort of query sequence |
| CN102254039A (en) * | 2011-08-11 | 2011-11-23 | 武汉安问科技发展有限责任公司 | Searching engine-based network searching method |
| CN102855252A (en) * | 2011-06-30 | 2013-01-02 | 北京百度网讯科技有限公司 | Method and device for data retrieval based on demands |
| CN102999625A (en) * | 2012-12-05 | 2013-03-27 | 北京海量融通软件技术有限公司 | Method for realizing semantic extension on retrieval request |
| CN103853722A (en) * | 2012-11-29 | 2014-06-11 | 腾讯科技（深圳）有限公司 | Query based keyword extension method, device and system |
| CN103914543A (en) * | 2014-04-03 | 2014-07-09 | 北京百度网讯科技有限公司 | Search result displaying method and device |
| CN104199822A (en) * | 2014-07-11 | 2014-12-10 | 五八同城信息技术有限公司 | Method and system for identifying demand classification corresponding to searching |
| US20140379745A1 (en) * | 2010-12-31 | 2014-12-25 | Alibaba Group Holding Limited | Recommendation of search keywords based on indication of user intention |
| CN105912630A (en) * | 2016-04-07 | 2016-08-31 | 北京搜狗科技发展有限公司 | Information expansion method and device |
| CN105975596A (en) * | 2016-05-10 | 2016-09-28 | 上海珍岛信息技术有限公司 | Query expansion method and system of search engine |
| CN106528588A (en) * | 2016-09-14 | 2017-03-22 | 厦门幻世网络科技有限公司 | Method and apparatus for matching resources for text information |
| CN106557476A (en) * | 2015-09-24 | 2017-04-05 | 北京奇虎科技有限公司 | The acquisition methods and device of relevant information |
| CN106610972A (en) * | 2015-10-21 | 2017-05-03 | 阿里巴巴集团控股有限公司 | Query rewriting method and apparatus |
| CN107544982A (en) * | 2016-06-24 | 2018-01-05 | 中兴通讯股份有限公司 | Text message processing method, device and terminal |
| CN107609152A (en) * | 2017-09-22 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | Method and apparatus for expanding query formula |
-
2018
- 2018-04-27 CN CN201810392779.4A patent/CN108776901B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102063468A (en) * | 2010-12-03 | 2011-05-18 | 百度在线网络技术（北京）有限公司 | Device and method for confirming query sort of query sequence |
| US20140379745A1 (en) * | 2010-12-31 | 2014-12-25 | Alibaba Group Holding Limited | Recommendation of search keywords based on indication of user intention |
| CN102855252A (en) * | 2011-06-30 | 2013-01-02 | 北京百度网讯科技有限公司 | Method and device for data retrieval based on demands |
| CN102254039A (en) * | 2011-08-11 | 2011-11-23 | 武汉安问科技发展有限责任公司 | Searching engine-based network searching method |
| CN103853722A (en) * | 2012-11-29 | 2014-06-11 | 腾讯科技（深圳）有限公司 | Query based keyword extension method, device and system |
| CN102999625A (en) * | 2012-12-05 | 2013-03-27 | 北京海量融通软件技术有限公司 | Method for realizing semantic extension on retrieval request |
| CN103914543A (en) * | 2014-04-03 | 2014-07-09 | 北京百度网讯科技有限公司 | Search result displaying method and device |
| CN104199822A (en) * | 2014-07-11 | 2014-12-10 | 五八同城信息技术有限公司 | Method and system for identifying demand classification corresponding to searching |
| CN106557476A (en) * | 2015-09-24 | 2017-04-05 | 北京奇虎科技有限公司 | The acquisition methods and device of relevant information |
| CN106610972A (en) * | 2015-10-21 | 2017-05-03 | 阿里巴巴集团控股有限公司 | Query rewriting method and apparatus |
| CN105912630A (en) * | 2016-04-07 | 2016-08-31 | 北京搜狗科技发展有限公司 | Information expansion method and device |
| CN105975596A (en) * | 2016-05-10 | 2016-09-28 | 上海珍岛信息技术有限公司 | Query expansion method and system of search engine |
| CN107544982A (en) * | 2016-06-24 | 2018-01-05 | 中兴通讯股份有限公司 | Text message processing method, device and terminal |
| CN106528588A (en) * | 2016-09-14 | 2017-03-22 | 厦门幻世网络科技有限公司 | Method and apparatus for matching resources for text information |
| CN107609152A (en) * | 2017-09-22 | 2018-01-19 | 百度在线网络技术（北京）有限公司 | Method and apparatus for expanding query formula |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111383069A (en) * | 2018-12-29 | 2020-07-07 | 美宅科技（北京）有限公司 | Furniture assembly recommendation method and device |
| CN109885753A (en) * | 2019-01-16 | 2019-06-14 | 苏宁易购集团股份有限公司 | A kind of method and device for expanding commercial articles searching and recalling |
| CN110222147A (en) * | 2019-05-15 | 2019-09-10 | 北京百度网讯科技有限公司 | Label extending method, device, computer equipment and storage medium |
| CN110781204A (en) * | 2019-09-09 | 2020-02-11 | 腾讯大地通途（北京）科技有限公司 | Identification information determination method, device, equipment and storage medium of target object |
| CN110781204B (en) * | 2019-09-09 | 2024-02-20 | 腾讯大地通途（北京）科技有限公司 | Identification information determining method, device, equipment and storage medium of target object |
| CN110706021A (en) * | 2019-09-12 | 2020-01-17 | 微梦创科网络科技（中国）有限公司 | Advertisement putting method and system |
| CN113111216A (en) * | 2020-01-13 | 2021-07-13 | 百度在线网络技术（北京）有限公司 | Advertisement recommendation method, device, equipment and storage medium |
| CN113111216B (en) * | 2020-01-13 | 2023-11-03 | 百度在线网络技术（北京）有限公司 | Advertisement recommendation method, device, equipment and storage medium |
| CN112256822A (en) * | 2020-10-21 | 2021-01-22 | 平安科技（深圳）有限公司 | Text search method and device, computer equipment and storage medium |
| CN112559693A (en) * | 2020-12-25 | 2021-03-26 | 鼎易创展咨询（北京）有限公司 | Similar word searching method, device, equipment and medium for text words |
| CN113297489A (en) * | 2021-05-28 | 2021-08-24 | 平安科技（深圳）有限公司 | Rehabilitation aid recommendation method and device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108776901B (en) | 2021-01-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108776901A (en) | Method and system for advertisement recommendation based on search term | |
| US11238081B2 (en) | Method, apparatus, and computer program product for classification and tagging of textual data | |
| US8533208B2 (en) | System and method for topic extraction and opinion mining | |
| CN109376309A (en) | Document recommendation method and device based on semantic label | |
| CN106663117B (en) | Constructing graphs supporting providing exploratory suggestions | |
| CN104462364B (en) | Method and device is recommended in search | |
| CN103870973B (en) | Information push, searching method and the device of keyword extraction based on electronic information | |
| CN109219811B (en) | Related paragraph retrieval system | |
| US20230177360A1 (en) | Surfacing unique facts for entities | |
| CN109829045A (en) | A kind of answering method and device | |
| CN109325146A (en) | A kind of video recommendation method, device, storage medium and server | |
| CN106599037A (en) | Recommendation method based on label semantic normalization | |
| CN110807326A (en) | Short text keyword extraction method combining GPU-DMM and text features | |
| Gu et al. | Service package recommendation for mashup creation via mashup textual description mining | |
| Kumar et al. | BERT based semi-supervised hybrid approach for aspect and sentiment classification | |
| CN109255012A (en) | A kind of machine reads the implementation method and device of understanding | |
| CN108062385A (en) | The method and system of Users' Interests Mining | |
| Zemlyanskiy et al. | DOCENT: Learning self-supervised entity representations from large document collections | |
| CN108536665A (en) | A kind of method and device of determining sentence consistency | |
| CN109284441A (en) | Dynamic self-adapting network sensitive information detection method and device | |
| Iftene et al. | Using semantic resources in image retrieval | |
| US20230376537A1 (en) | Multi-chunk relationship extraction and maximization of query answer coherence | |
| CN115618871A (en) | Merchant text identification method, device, equipment and storage medium | |
| Botana et al. | Explain and conquer: Personalised text-based reviews to achieve transparency | |
| CN114077834A (en) | Method, device and storage medium for determining similar texts |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |