CN108628981A - A kind of article method for pushing and system based on body index - Google Patents
A kind of article method for pushing and system based on body index Download PDFInfo
- Publication number
- CN108628981A CN108628981A CN201810389652.7A CN201810389652A CN108628981A CN 108628981 A CN108628981 A CN 108628981A CN 201810389652 A CN201810389652 A CN 201810389652A CN 108628981 A CN108628981 A CN 108628981A
- Authority
- CN
- China
- Prior art keywords
- article
- default
- vector
- label
- body index
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
- G06F40/186—Templates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of article method for pushing based on body index, includes the following steps：S2000 is marked the article in article library according to default classifying rules；Each article corresponds to one or more default labels；S3000 obtains the body index information of user, and the body index label of user is generated according to the body index information of user；S4000 matches the body index label with default label, searches the target being consistent with the body index label and presets label, presets the article of label labeled as the target to user's push.The present invention pushes corresponding article by the body index of user to user, when user wants to understand the physical condition of oneself and improve the method for physical condition, corresponding information can easily be obtained, and often to user's push and the relevant article of oneself physical condition, user can be reminded to take exercise, to improve the physical condition of user.
Description
Technical field
The invention belongs to content push technology field, more particularly to a kind of article method for pushing based on body index and it is
System.
Background technology
With the development of Internet technology, huge change also has occurred in people’s lives mode.People are increasingly
More obtains information by internet, and the process of information is obtained to shorten user by internet, and website or application are past
It actively recommend some contents to user toward meeting.For moving the method for pushing of class article, movement class article method for pushing is main
It is the content tab by all stamping the types such as content tab, such as keyword, theme, classification to every movement class article, then
It browses according to the behavioural information of user, such as user, which movement class article was read, by the method for machine learning to user
Corresponding user tag, such as the user tag of keyword and theme type are stamped, when user generates browse request, for example is used
When the application interface of family open movement class website or drop-down movement class application, according to the user tag of user, in movement class article
The corresponding movement class article of user tag that user is retrieved in library, is pushed to user.And based on the behavioural information of user come
It is pushed, the user demand obtained is inaccurate, and the article for being susceptible to push is not the information that user wants, and is drawn
The dislike for playing user, reduces the Experience Degree of user.
Invention content
The object of the present invention is to provide a kind of article method for pushing and system based on body index, realization meet user's need
It asks, improves the purpose of user experience.
Technical solution provided by the invention is as follows：
A kind of article method for pushing based on body index, includes the following steps：
S2000 is marked the article in article library according to default classifying rules；Each article corresponds to one or more
Default label；
S3000 obtains the body index information of user, and the body index of user is generated according to the body index information of user
Label；
S4000 matches the body index label with default label, and lookup is consistent with the body index label
Target preset label, to user push labeled as the target preset label article.
Further, the step S2000 is specifically included：
S2100 obtains the default article template of current preset label；
S2200 calculates the similarity of the article and the current preset article template in article library；
S2300 judges whether the similarity is more than the corresponding default similarity of the current preset label, if so, holding
Row step S2400；
S2400 marks the article according to the current preset label.
Further, the step S2200 is specifically included：
S2210 segments the default article template, generates and presets term vector；
S2220 segments the article in article library, generates article term vector；
S2230 generated according to the default term vector and the article term vector preset word frequency vector sum article word frequency to
Amount；
S2240 calculates the similarity of the default word frequency vector and the article word frequency vector.
Further, the step S2230 is specifically included：
S2231 calculates the weighted value of all words in the default term vector, and according to the big of the weighted value of all words
The small keyword for choosing predetermined number generates predetermined keyword vector；
S2232 calculates the weighted value of all words in the article term vector, and according to the big of the weighted value of all words
The small keyword for choosing predetermined number generates article key term vector；
The predetermined keyword vector sum article key term vector union is generated total crucial term vector by S2233；
S2234 calculates word frequency of the keyword in the predetermined keyword vector in total crucial term vector, and root
The default word frequency vector is generated according to result of calculation；
S2235 calculates word frequency of the keyword in the article key term vector in total crucial term vector, and root
The article word frequency vector is generated according to result of calculation.
Further, further include before the step S2000：
S1000, which is obtained, presets body index information, and body index tag library is generated according to default body index information；It is described
Body index tag library includes multiple and different default label.
The present invention also provides a kind of article supplying system based on body index, including：
Mark module, for the article in article library to be marked according to default classifying rules；Each article corresponds to one
A or multiple default labels；
First generation module, the body index information for obtaining user, and generated according to the body index information of user
The body index label of user；
Pushing module is searched and the body index for matching the body index label with default label
The target that label is consistent presets label, presets the article of label labeled as the target to user's push.
Further, the mark module includes：
Acquisition submodule, the default article template for obtaining current preset label；
Computational submodule calculates the similarity of the article and the current preset article template in article library；
Judging submodule, the default similarity whether being more than in the current preset label for judging the similarity；
Implementation sub-module, for judging when the similarity is more than the default similarity in the current preset label,
The article is marked according to the current preset label.
Further, the computational submodule includes：
First participle unit generates for being segmented to default article template and presets term vector；
Second participle unit generates article term vector for being segmented to the article in article library；
Vectorial generation unit presets word frequency vector sum for being generated according to the default term vector and the article term vector
Article word frequency vector；
Computing unit, the similarity for calculating the default word frequency vector and the article word frequency vector.
Further, the vectorial generation unit includes：
Primary vector generation subelement, the weighted value for calculating all words in the default term vector, and according to institute
There is the keyword that the size of the weighted value of word chooses predetermined number to generate predetermined keyword vector；
Secondary vector generation subelement, the weighted value for calculating all words in the article term vector, and according to institute
There is the keyword that the size of the weighted value of word chooses predetermined number to generate article key term vector；
Third vector generates subelement, for by the predetermined keyword vector sum article key term vector union, generating
Total key term vector；
4th vector generates subelement, for calculating the keyword in the predetermined keyword vector in total keyword
Word frequency in vector, and the default word frequency vector is generated according to result of calculation；
5th vector generates subelement, for calculating the keyword in the article key term vector in total keyword
Word frequency in vector, and the article word frequency vector is generated according to result of calculation.
Further, the system also includes the second generation modules, for obtaining default body index information, according to default
Body index information generates body index tag library；The body index tag library includes multiple and different default label.
A kind of the article method for pushing and system based on body index provided through the invention, can bring it is following at least
A kind of advantageous effect：
1, the present invention is by obtaining the body index information of user and generating the body index label of user, then according to
The body index at family pushes corresponding article to user, when user wants to understand the physical condition of oneself and improves the side of physical condition
When method, you can easily obtain corresponding information.
2, the present invention is by that often to user's push and the relevant article of its physical condition, can remind user to carry out sport forging
Refining, to improve the physical condition of user.
3, the body index for the user that the present invention obtains is dynamic, when the physical condition of user improves, such as by partially fat
When type becomes standard type, then to the article of the push of user from becoming the text labeled as standard type labeled as the article of inclined fat type
Chapter can more meet the demand of user, will not cause the dislike of user, and the Experience Degree of user can be improved.
4, the present invention marks the article in article library by default article template so that label accuracy rate higher, to
Keep the article information that user obtains more acurrate, improves the Experience Degree of user.
Description of the drawings
Below by a manner of clearly understandable, preferred embodiment is described with reference to the drawings, to a kind of above-mentioned characteristic, technology
Feature, advantage and its realization method are further described.
Fig. 1 is a kind of flow diagram of article method for pushing based on body index of the embodiment of the present invention one；
Fig. 2 is a kind of flow diagram of article method for pushing based on body index of the embodiment of the present invention two；
Fig. 3 is a kind of flow diagram of article method for pushing based on body index of the embodiment of the present invention three；
Fig. 4 is a kind of flow diagram of article method for pushing based on body index of the embodiment of the present invention four；
Fig. 5 is a kind of structural schematic block diagram of article supplying system based on body index of the embodiment of the present invention five.
Drawing reference numeral explanation：
10, the second generation module；20, mark module；21, acquisition submodule；22, computational submodule；220, the first participle
Unit；230, the second participle unit；240, vectorial generation unit；241, primary vector generates subelement；242, secondary vector is given birth to
At subelement；243, third vector generates subelement；244, the 4th vector generates subelement；245, it is single to generate son for the 5th vector
Member；250, computing unit；23, judging submodule；24, implementation sub-module；30, the first generation module；40, pushing module.
Specific implementation mode
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, control is illustrated below
The specific implementation mode of the present invention.It should be evident that drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, other are can also be obtained according to these attached drawings
Attached drawing, and obtain other embodiments.
To make simplified form, part related to the present invention is only schematically shown in each figure, they are not represented
Its practical structures as product.In addition, so that simplified form is easy to understand, there is identical structure or function in some figures
Component only symbolically depicts one of those, or has only marked one of those.Herein, "one" is not only indicated
" only this ", can also indicate the situation of " more than one ".
According to first embodiment provided by the invention, as shown in Figure 1,
A kind of article method for pushing based on body index, includes the following steps：
S2000 is marked the article in article library according to default classifying rules；Each article corresponds to one or more
Default label；
S3000 obtains the body index information of user, and the body index of user is generated according to the body index information of user
Label；
S4000 matches the body index label with default label, and lookup is consistent with the body index label
Target preset label, to user push labeled as the target preset label article.
Specifically, in daily life, user can pass through the interested article of a variety of application searches oneself of terminal device
Or the article of a variety of application push is read in selection.Above application can be the social software comprising information flow advertising function or various
Information service platform etc..The present invention is first by the article in social software or information service platform article library according to default classifying rules
It is marked so that each article corresponds to one or more labels；Then the body index information of user, the body of user are obtained
Indication information includes but not limited to the gender of user, age, BMI, body fat, internal organ body fat, muscle, moisture, bone, basic generation
It thanks, protein, heart rate, blood glucose and blood pressure etc..The body index information of user can be claimed by intelligent body fat, fatty measuring instrument, blood
The measurements such as sugar meter and electronic sphygmomanometer obtain.After terminal device gets the body index information of user, according to the body of user
Indication information generates the body index label of user, such as according to the BMI of user and body fat index, the body index of the user of generation
Label is inclined fat type, then matches the body index label of user (inclined fat type) with default label, is searched and user's body
The target that body index label (inclined fat type) is consistent presets label, i.e., the label for being preset as inclined fat type is searched in default label, so
Preset the article of label (inclined fat type) labeled as target to user's push again afterwards.The present invention by obtain user body index,
And corresponding article is pushed to user according to the body index of user, when user wants to understand the physical condition of oneself and improves body
When the method for situation, you can easily obtain corresponding information, and often relevant with oneself physical condition to user's push
Article can remind user to take exercise, to improve the physical condition of user, meanwhile, the body index of user be it is dynamic,
When the physical condition of user improves, when such as becoming standard type from inclined fat type, then to the article of the push of user by being labeled as partially
The article of fat type becomes the article labeled as standard type, can more meet the demand of user, will not cause the dislike of user, improves and uses
The Experience Degree at family.
According to second embodiment provided by the invention, as shown in Fig. 2,
A kind of article method for pushing based on body index, includes the following steps：
S2100 obtains the default article template of current preset label；
S2200 calculates the similarity of the article and the current preset article template in article library；
S2300 judges whether the similarity is more than the corresponding default similarity of the current preset label, if so, holding
Row step S2400；
S2400 marks the article according to the current preset label；
S3000 obtains the body index information of user, and the body index of user is generated according to the body index information of user
Label；
S4000 matches the body index label with default label, and lookup is consistent with the body index label
Target preset label, to user push labeled as the target preset label article.
Specifically, label preset to each, it is special first to meet corresponding default label by hand picking one in article library
A default article template is arranged in the article of sign, as each default label.Then by article library article with it is all pre-
The default article template of bidding label is compared, and calculates the default article template of the article in article library and all default labels
Similarity, when a certain article reaches corresponding default similarity with the similarity of the default article template of a certain default label
When, then by this article labeled as the default label.The above method is recycled, all articles in article library are marked.
For example, current article to be sorted is A, multiple default labels are respectively B, C, D, preset the default phase of label B, C, D
It is respectively 90%, 85%, 80% like degree.First respectively default label B, C, D select a default article template, then count respectively
It calculates article A and the similarity of the default article template of default label B, article A is similar to the default article template of default label C
The similarity of the default article template of degree, article A and default label D, such as the default article template of article A and default label B
Similarity is more than 90%, then article A is labeled as default label B；Such as the phase of article A and the default article template of default label C
Like degree also greater than 85%, then article A is labeled as default label C simultaneously；I.e. by article A while labeled as default label B and in advance
If label C；If the similarity of article A and the default article template of default label D is also greater than 80%, then article A is marked simultaneously
For default label B, C, D.Such as the both less than default label pair of the similarity of the default article template of article A and default label B, C, D
The default similarity answered, then be not marked article A.The present invention marks the text in article library by default article template
Chapter so that label accuracy rate higher, the article information obtained thereby using family is more acurrate, can more meet the use demand of user,
Improve the Experience Degree of user.
According to 3rd embodiment provided by the invention, as shown in figure 3,
A kind of article method for pushing based on body index, includes the following steps：
S2100 obtains the default article template of current preset label；
S2210 segments the default article template, generates and presets term vector；
S2220 segments the article in article library, generates article term vector；
S2231 calculates the weighted value of all words in the default term vector, and according to the big of the weighted value of all words
The small keyword for choosing predetermined number generates predetermined keyword vector；
S2232 calculates the weighted value of all words in the article term vector, and according to the big of the weighted value of all words
The small keyword for choosing predetermined number generates article key term vector；
The predetermined keyword vector sum article key term vector union is generated total crucial term vector by S2233；
S2234 calculates word frequency of the keyword in the predetermined keyword vector in total crucial term vector, and root
The default word frequency vector is generated according to result of calculation；
S2235 calculates word frequency of the keyword in the article key term vector in total crucial term vector, and root
The article word frequency vector is generated according to result of calculation；
S2240 calculates the similarity of the default word frequency vector and the article word frequency vector；
S2300 judges whether the similarity is more than the corresponding default similarity of the current preset label, if so, holding
Row step S2400；
S2400 marks the article according to the current preset label；
S3000 obtains the body index information of user, and the body index of user is generated according to the body index information of user
Label；
S4000 matches the body index label with default label, and lookup is consistent with the body index label
Target preset label, to user push labeled as the target preset label article.
Specifically, be after each default label select presets article template calculate separately article in article library with
The similarity of the default article template of all default labels, this sentences the default article template of an article and a default label
Similarity calculating for illustrate.
Default article template is segmented first, the stop-word that cannot reflect article content feature is rejected, such as rejects text
In chapter " ", "Yes", " " etc., generate the default term vector for presetting article template, preset term vector=[word n1, word
n2, word n3... word ni], then i ∈ N segment article to be marked, generate the article word of article to be marked to
Amount, article term vector=[word t1, word t2, word t3... word tj], j ∈ N.It calculates and presets term vector and article word
The weighted value of all words in vector.
The weighted value for calculating all words in default term vector and article term vector is described as follows；
The TF-IDF weighted values for presetting all words in term vector and article term vector are calculated first；Wherein, word is normalized
Frequently (TF)=[F (word ni) ÷ F (max)], F (word ni) i.e. word niThe frequency occurred in default article template, F (max)
The frequency that all words in i.e. default term vector occur in default article template,；Preferably, F (max) refers in default text
The frequency of the maximum word of the frequency of occurrences, can simplify calculating in this way in chapter template, promote the computational efficiency of weighted value.
Inverse document frequency (IDF)=log [in article library article sum/(include word niArticle number+1)]；
TF-IDF weighted values=normalization word frequency (TF) × inverse document frequency (IDF).
It calculates in default term vector and article term vector after the TF-IDF weighted values of all words, it will be in default term vector
All words descending arrangement is carried out according to the size of TF-IDF weighted values, take the word of front m1 as presetting article template
Keyword, and generate predetermined keyword vector, predetermined keyword vector=[keyword n1, keyword n2... keyword
nm1].All words in article term vector are similarly subjected to descending arrangement according to the size of TF-IDF weighted values, take front m2
Keyword of the word as article to be marked, and article key term vector is generated, article key term vector=[keyword t1, close
Keyword t2... keyword tm2].Wherein, m1 can be equal with m2, also can be unequal.
M1 and m2 is unequal in order to prevent, the present invention by by predetermined keyword vector sum article key term vector union,
Then word frequency of the keyword in predetermined keyword vector sum article key term vector for this union is calculated separately, and is generated
Default word frequency vector and article word frequency vector, the element number phase of the default word frequency vector obtained at this time and article word frequency vector
Together, the similarity for presetting article template and article template to be marked is calculated to facilitate.For example, predetermined keyword it is vectorial=[A, B,
C, D], article key term vector=[A, C, D, E, F], total key term vector=[A, B, C, D, E, F] after union preset key
Word frequency of the keyword in total crucial term vector in term vector is respectively 1,1,1,1,0,0, i.e., the word frequency of keyword A is 1, and
There is no E in predetermined keyword vector, then the word frequency of E is 0, and the default word frequency vector a=[1,1,1,1,0,0] of generation is similarly given birth to
At article word frequency vector b=[1,0,1,1,1,1], finally calculate preset word frequency vector it is similar to the cosine of article word frequency vector
Degree, wherein cosine similarity Calculate the cosine phase of article to be marked and default article template
After degree, judge whether the cosine similarity is more than the default similarity for presetting the corresponding default label of article template, if so,
Label is then preset according to this and marks the article to be marked, if it is not, the article to be marked is not marked then.The present invention passes through
Calculate the TF-IDF weighted values for presetting all words in article template and article to be marked, then calculate again default article template with
The cosine similarity of article to be marked, to determine whether current article to be marked is labeled as current preset label so that label
Accuracy rate higher can more meet the use demand of user, improve the Experience Degree of user.
Fourth embodiment provided by the invention, as shown in figure 4,
A kind of article method for pushing based on body index, includes the following steps：
S1000, which is obtained, presets body index information, and body index tag library is generated according to default body index information；It is described
Body index tag library includes multiple and different default label；
S2100 obtains the default article template of current preset label；
S2210 segments the default article template, generates and presets term vector；
S2220 segments the article in article library, generates article term vector；
S2231 calculates the weighted value of all words in the default term vector, and according to the big of the weighted value of all words
The small keyword for choosing predetermined number generates predetermined keyword vector；
S2232 calculates the weighted value of all words in the article term vector, and according to the big of the weighted value of all words
The small keyword for choosing predetermined number generates article key term vector；
The predetermined keyword vector sum article key term vector union is generated total crucial term vector by S2233；
S2234 calculates word frequency of the keyword in the predetermined keyword vector in total crucial term vector, and root
The default word frequency vector is generated according to result of calculation；
S2235 calculates word frequency of the keyword in the article key term vector in total crucial term vector, and root
The article word frequency vector is generated according to result of calculation；
S2240 calculates the similarity of the default word frequency vector and the article word frequency vector；
S2300 judges whether the similarity is more than the corresponding default similarity of the current preset label, if so, holding
Row step S2400；
S2400 marks the article according to the current preset label；
S3000 obtains the body index information of user, and the body index of user is generated according to the body index information of user
Label；
S4000 matches the body index label with default label, and lookup is consistent with the body index label
Target preset label, to user push labeled as the target preset label article.
Specifically, preset body index information include but not limited to gender, the age, BMI, body fat, internal organ body fat, muscle,
Moisture, bone, basic metabolism, protein, heart rate, blood glucose and blood pressure etc..According to different default body index information, produce
Different body index tag libraries.Such as by being associated with two body indexs of BMI and body fat, the body index tag library of generation can wrap
Include Marasmus Malnutrition, inclined thin, the slender type of muscle, slender type, health type, muscularity, sportsman's type, cryptomorphic obesity type, inclined fat type, fertilizer
The physical conditions labels such as the builds such as fat type label and interior fat are higher, muscle is relatively low, bone amount is relatively low.Certainly, actually make
Used time can also generate different body index tag libraries by being associated with different body indexs.
A default article template is selected for all default labels in body index tag library, is then calculated in article library
The similarity of article and all default article templates will if similarity meets the default similarity of corresponding default label
This article is labeled as the default label, and an article can correspond to one or more default labels, by all articles in article library
It is marked using the above method.After having marked, index is established for the article of default label to all labels, it is soft to improve
Part search quickly pushes corresponding article convenient for software labeled as the speed of the article of same default label to user.
After getting the body index information of user, the body index of user is generated according to the body index information of user
Then label searches the target being consistent with user's body index label in body index tag library and presets label, then will be literary
Label is pushed to user for the article of default label in Zhang Ku.The present invention is pushed by the body index of user to user
Article can not only help user to understand the physical condition of oneself, but also user can be supervised to carry out physical exercise, to improve user's
Fitness.
5th embodiment provided by the invention, as shown in figure 5,
A kind of article supplying system based on body index, including：
Mark module 20, for the article in article library to be marked according to default classifying rules；Each article corresponds to
The default label of one or more；
First generation module 30, the body index information for obtaining user, and given birth to according to the body index information of user
At the body index label of user；
Pushing module 40, for matching the body index label with default label, lookup refers to the body
The target that mark label is consistent presets label, presets the article of label labeled as the target to user's push.
Preferably, mark module 20 includes：Acquisition submodule 21, the default article mould for obtaining current preset label
Plate；
Computational submodule 22 calculates the similarity of the article and the current preset article template in article library；
Judging submodule 23, for judging it is default similar in the current preset label whether the similarity is more than
Degree；
Implementation sub-module 24 is more than the default similarity in the current preset label for judging when the similarity
When, the article is marked according to the current preset label.
Preferably, computational submodule 22 includes：
First participle unit 220 generates for being segmented to default article template and presets term vector；
Second participle unit 230 generates article term vector for being segmented to the article in article library；
Vectorial generation unit 240, for generated according to the default term vector and the article term vector preset word frequency to
Amount and article word frequency vector；
Computing unit 250, the similarity for calculating the default word frequency vector and the article word frequency vector.
Preferably, vectorial generation unit 240 includes：
Primary vector generation subelement 241, the weighted value for calculating all words in the default term vector, and according to
The keyword that the size of the weighted value of all words chooses predetermined number generates predetermined keyword vector；
Secondary vector generation subelement 242, the weighted value for calculating all words in the article term vector, and according to
The keyword that the size of the weighted value of all words chooses predetermined number generates article key term vector；
Third vector generates subelement 243, is used for the predetermined keyword vector sum article key term vector union, raw
At total crucial term vector；
4th vector generates subelement 244, for calculating the keyword in the predetermined keyword vector in total pass
Word frequency in keyword vector, and the default word frequency vector is generated according to result of calculation；
5th vector generates subelement 245, for calculating the keyword in the article key term vector in total pass
Word frequency in keyword vector, and the article word frequency vector is generated according to result of calculation.
Preferably, system further includes the second generation module 10, for obtaining default body index information, according to default body
Indication information generates body index tag library；The body index tag library includes multiple and different default label.
The concrete mode that modules in the present embodiment execute operation carries out in the embodiment of the method
Detailed description, will be not set forth in detail explanation herein.
It should be noted that above-described embodiment can be freely combined as needed.The above is only the preferred of the present invention
Embodiment, it is noted that for those skilled in the art, in the premise for not departing from the principle of the invention
Under, several improvements and modifications can also be made, these improvements and modifications also should be regarded as protection scope of the present invention.
Claims (10)
1. a kind of article method for pushing based on body index, which is characterized in that include the following steps：
S2000 is marked the article in article library according to default classifying rules；Each article corresponds to one or more default
Label；
S3000 obtains the body index information of user, and the body index label of user is generated according to the body index information of user；
S4000 matches the body index label with default label, searches the mesh being consistent with the body index label
Mark presets label, presets the article of label labeled as the target to user's push.
2. a kind of article method for pushing based on body index according to claim 1, which is characterized in that the step
S2000 is specifically included：
S2100 obtains the default article template of current preset label；
S2200 calculates the similarity of the article and the current preset article template in article library；
S2300 judges whether the similarity is more than the corresponding default similarity of the current preset label, if so, executing step
Rapid S2400；
S2400 marks the article according to the current preset label.
3. a kind of article method for pushing based on body index according to claim 2, which is characterized in that the step
S2200 is specifically included：
S2210 segments the default article template, generates and presets term vector；
S2220 segments the article in article library, generates article term vector；
S2230 is generated according to the default term vector and the article term vector presets word frequency vector sum article word frequency vector；
S2240 calculates the similarity of the default word frequency vector and the article word frequency vector.
4. a kind of article method for pushing based on body index according to claim 3, which is characterized in that the step
S2230 is specifically included：
S2231 calculates the weighted value of all words in the default term vector, and is selected according to the size of the weighted value of all words
The keyword of predetermined number is taken to generate predetermined keyword vector；
S2232 calculates the weighted value of all words in the article term vector, and is selected according to the size of the weighted value of all words
The keyword of predetermined number is taken to generate article key term vector；
The predetermined keyword vector sum article key term vector union is generated total crucial term vector by S2233；
S2234 calculates word frequency of the keyword in the predetermined keyword vector in total crucial term vector, and according to meter
It calculates result and generates the default word frequency vector；
S2235 calculates word frequency of the keyword in the article key term vector in total crucial term vector, and according to meter
It calculates result and generates the article word frequency vector.
5. a kind of article method for pushing based on body index according to claim 1, which is characterized in that the step
Further include before S2000：
S1000, which is obtained, presets body index information, and body index tag library is generated according to default body index information；The body
Index tag library includes multiple and different default label.
6. a kind of article supplying system based on body index, which is characterized in that including：
Mark module, for the article in article library to be marked according to default classifying rules；Each article correspond to one or
Multiple default labels；
First generation module, the body index information for obtaining user, and user is generated according to the body index information of user
Body index label；
Pushing module is searched and the body index label for matching the body index label with default label
The target being consistent presets label, presets the article of label labeled as the target to user's push.
7. a kind of article supplying system based on body index according to claim 6, which is characterized in that the label mould
Block includes：
Acquisition submodule, the default article template for obtaining current preset label；
Computational submodule calculates the similarity of the article and the current preset article template in article library；
Judging submodule, the default similarity whether being more than in the current preset label for judging the similarity；
Implementation sub-module, for judging when the similarity is more than the default similarity in the current preset label, according to
The current preset label marks the article.
8. a kind of article supplying system based on body index according to claim 7, which is characterized in that calculating
Module includes：
First participle unit generates for being segmented to default article template and presets term vector；
Second participle unit generates article term vector for being segmented to the article in article library；
Vectorial generation unit presets word frequency vector sum article for being generated according to the default term vector and the article term vector
Word frequency vector；
Computing unit, the similarity for calculating the default word frequency vector and the article word frequency vector.
9. a kind of article supplying system based on body index according to claim 8, which is characterized in that the vector is raw
Include at unit：
Primary vector generation subelement, the weighted value for calculating all words in the default term vector, and according to all words
The keyword that the size of the weighted value of language chooses predetermined number generates predetermined keyword vector；
Secondary vector generation subelement, the weighted value for calculating all words in the article term vector, and according to all words
The keyword that the size of the weighted value of language chooses predetermined number generates article key term vector；
Third vector generates subelement, for by the predetermined keyword vector sum article key term vector union, generating total close
Keyword vector；
4th vector generates subelement, for calculating the keyword in the predetermined keyword vector in total crucial term vector
In word frequency, and the default word frequency vector is generated according to result of calculation；
5th vector generates subelement, for calculating the keyword in the article key term vector in total crucial term vector
In word frequency, and article word frequency vector is generated according to result of calculation.
10. a kind of article supplying system based on body index according to claim 6, which is characterized in that
The system also includes the second generation modules, for obtaining default body index information, according to default body index information
Generate body index tag library；The body index tag library includes multiple and different default label.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810389652.7A CN108628981A (en) | 2018-04-27 | 2018-04-27 | A kind of article method for pushing and system based on body index |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810389652.7A CN108628981A (en) | 2018-04-27 | 2018-04-27 | A kind of article method for pushing and system based on body index |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108628981A true CN108628981A (en) | 2018-10-09 |
Family
ID=63694745
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810389652.7A Pending CN108628981A (en) | 2018-04-27 | 2018-04-27 | A kind of article method for pushing and system based on body index |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108628981A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111125566A (en) * | 2019-12-11 | 2020-05-08 | 贝壳技术有限公司 | Information acquisition method and device, electronic equipment and storage medium |
| CN111916225A (en) * | 2020-07-31 | 2020-11-10 | 东南大学 | Android platform-based blood pressure health data management system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102467609A (en) * | 2010-10-29 | 2012-05-23 | 大荣优比泰克有限公司 | Media recommending system based on health index |
| CN103778171A (en) * | 2012-10-24 | 2014-05-07 | I-昂索夫特有限公司 | System for health media recommendation based on wellbeing-index |
| WO2014107874A1 (en) * | 2013-01-11 | 2014-07-17 | Nokia Corporation | Method and apparatus for enriching social media to improve personalized user experience |
| CN106156462A (en) * | 2015-04-01 | 2016-11-23 | 时云医疗科技（上海）有限公司 | A kind of behavioural information method for pushing based on sign data and system |
| KR101733230B1 (en) * | 2015-05-21 | 2017-05-08 | 네이버 주식회사 | Method for providing contents sharing service, method for controlling display, and computer program for executing one of the method |
| CN106649274A (en) * | 2016-12-27 | 2017-05-10 | 东华互联宜家数据服务有限公司 | Text content tag labeling method and device |
| CN107222570A (en) * | 2017-07-10 | 2017-09-29 | 江西博瑞彤芸科技有限公司 | A kind of health information method for pushing based on user's behaviour in service |
-
2018
- 2018-04-27 CN CN201810389652.7A patent/CN108628981A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102467609A (en) * | 2010-10-29 | 2012-05-23 | 大荣优比泰克有限公司 | Media recommending system based on health index |
| CN103778171A (en) * | 2012-10-24 | 2014-05-07 | I-昂索夫特有限公司 | System for health media recommendation based on wellbeing-index |
| WO2014107874A1 (en) * | 2013-01-11 | 2014-07-17 | Nokia Corporation | Method and apparatus for enriching social media to improve personalized user experience |
| CN106156462A (en) * | 2015-04-01 | 2016-11-23 | 时云医疗科技（上海）有限公司 | A kind of behavioural information method for pushing based on sign data and system |
| KR101733230B1 (en) * | 2015-05-21 | 2017-05-08 | 네이버 주식회사 | Method for providing contents sharing service, method for controlling display, and computer program for executing one of the method |
| CN106649274A (en) * | 2016-12-27 | 2017-05-10 | 东华互联宜家数据服务有限公司 | Text content tag labeling method and device |
| CN107222570A (en) * | 2017-07-10 | 2017-09-29 | 江西博瑞彤芸科技有限公司 | A kind of health information method for pushing based on user's behaviour in service |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111125566A (en) * | 2019-12-11 | 2020-05-08 | 贝壳技术有限公司 | Information acquisition method and device, electronic equipment and storage medium |
| CN111125566B (en) * | 2019-12-11 | 2021-08-31 | 贝壳找房(北京)科技有限公司 | Information acquisition method and device, electronic equipment and storage medium |
| CN111916225A (en) * | 2020-07-31 | 2020-11-10 | 东南大学 | Android platform-based blood pressure health data management system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11113714B2 (en) | Filtering machine for sponsored content | |
| US11907277B2 (en) | Method, apparatus, and computer program product for classification and tagging of textual data | |
| CN105653562B (en) | The calculation method and device of correlation between a kind of content of text and inquiry request | |
| Wu et al. | Learning to tag | |
| US20230316377A1 (en) | Methods, systems, and media for providing information based on grouping information | |
| US8856129B2 (en) | Flexible and scalable structured web data extraction | |
| CN108334592B (en) | Personalized recommendation method based on combination of content and collaborative filtering | |
| CN110147499B (en) | Labeling method, recommendation method and recording medium | |
| CN106529158B (en) | The method and device of recommended dietary information | |
| JP6033697B2 (en) | Image evaluation device | |
| Wei et al. | Exploring heterogeneous features for query-focused summarization of categorized community answers | |
| Yang et al. | Modeling user interests for zero-query ranking | |
| CN108628981A (en) | A kind of article method for pushing and system based on body index | |
| Starke et al. | Promoting healthy food choices online: A case for multi-list recommender systems | |
| CN103729431B (en) | Massive microblog data distributed classification device and method with increment and decrement function | |
| CN110853728A (en) | Health recommendation information processing system and method | |
| CN106971004A (en) | Search result provides method and device | |
| Kumar et al. | Food calorie estimation system using imageai with retinanet feature extraction | |
| López-Iñesta et al. | Classification similarity learning using feature-based and distance-based representations: A comparative study | |
| DiBartola et al. | Metrics and the scientific literature: deciding what to read | |
| CN111292151B (en) | Object recommendation method and device, storage medium and object recommendation equipment | |
| Cherednichenko et al. | Item Matching Model in E-Commerce: How Users Benefit | |
| Ochi et al. | Rating prediction by correcting user rating bias | |
| Wu et al. | UESTC at ImageCLEF 2010 Medical Retrieval Task. | |
| WO2023079703A1 (en) | Processing execution system, processing execution method, and program |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181009 |
|
| WD01 | Invention patent application deemed withdrawn after publication |