CN106897369B - Content data recommendation method and system - Google Patents
Content data recommendation method and system Download PDFInfo
- Publication number
- CN106897369B CN106897369B CN201710031491.XA CN201710031491A CN106897369B CN 106897369 B CN106897369 B CN 106897369B CN 201710031491 A CN201710031491 A CN 201710031491A CN 106897369 B CN106897369 B CN 106897369B
- Authority
- CN
- China
- Prior art keywords
- data
- probability
- user
- check
- preset
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (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 relates to a content data recommendation method and a content data recommendation system, belongs to the field of content data recommendation, and can accurately push proper content data to a user according to the real-time condition of the user without spending a large amount of time to collect historical data of the user, so that the receiving and using degree of the pushed content data by the user is improved. When a user uses the movable equipment, the physiological data of the user and the action data of shaking the movable equipment are obtained through the sensor on the movable equipment; processing the acquired physiological data and action data of the user to obtain characteristic data of the user; and obtaining content data matched with the characteristic data of the user according to a preset content tag in a content database associated with the mobile equipment, and pushing the content data to the user. The method and the device are used for realizing accurate recommendation of content data to the user.
Description
Technical Field
The present invention relates to the field of content data recommendation.
Background
In the prior art, data recommendation generally includes ways of pushing data randomly, pushing data according to a history record, pushing data fixedly, and the like. Random data pushing, namely, randomly selecting data from a data database and pushing the data to a user, wherein most of the data recommended by the method cannot meet the current use of the user, and the accuracy is poor. In this way, although the pushed content data is accurate, it is necessary to accumulate the usage data of the user for a long time, and it takes a lot of time to collect the data usage of the user. The method comprises the steps of pushing fixed data to a user according to actual conditions, manually setting the fixed data, recommending content data through the method, not automatically pushing according to user characteristics, and enabling the user to have low acceptance utilization rate of the pushed data.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a content data recommendation method and system, which can accurately push appropriate content data to a user according to the real-time situation of the user, without spending a large amount of time to collect historical data of the user, and improve the receiving and using degree of the pushed content data by the user.
The technical scheme for solving the technical problems is as follows:
a content data recommendation method comprising the steps of:
s1, when the user uses the mobile device, acquiring physiological data of the user and motion data of shaking the mobile device through a sensor on the mobile device;
s2, processing the acquired physiological data and the acquired action data of the user to obtain characteristic data of the user;
and S3, obtaining content data matched with the characteristic data of the user according to a preset content tag in a content database associated with the mobile device, and pushing the content data to the user.
The invention has the beneficial effects that: the method and the device can accurately push the appropriate content data to the user according to the real-time situation of the user, do not need to spend a large amount of time on collecting the historical data of the user, and improve the receiving and using degree of the pushed content data of the user.
On the basis of the technical scheme, the invention can be further improved as follows:
further, in S1, the physiological data is body temperature, and the motion data includes: the motion strength, frequency, speed, amplitude and direction; the characteristic data of the user comprises: sex and age.
The beneficial effect of adopting the further scheme is that: the gender and the age of the user are judged by collecting the body temperature and the action data of the user, the characteristic data of the user can be obtained in real time, and the content data of the composite user characteristic can be accurately pushed.
Further, in S2, the processing to obtain the feature data of the user includes:
s21, obtaining an initial value of first probability data indicating a gender of the user, an initial value of second probability data indicating an age of the user, and initial values of a plurality of pieces of check data for correcting the first probability data and the second probability data, based on the physiological data and the motion data;
s22, adjusting the first probability data according to the plurality of check data, updating the second probability data and the plurality of check data according to a preset formula associated with the first probability data corresponding to the second probability data and the plurality of check data, respectively, so that the plurality of check data are all in a preset range corresponding to the second probability data and the plurality of check data are all in a preset target range corresponding to the first probability data and the second probability data;
and S23, obtaining the characteristic data of the user according to the range of the values of the first probability data and the second probability data.
The beneficial effect of adopting the further scheme is that: the probability that the user is different in gender and age is corrected through the range of the plurality of check data, errors caused by judging the gender and the age of the user according to single data are avoided, and the accuracy of judging the gender and the age of the user is greatly improved.
Further, in S21, the plurality of check data includes: the first check data, the second check data, the third check data and the fourth check data; the S22 further includes:
s221, updating second probability data according to a preset section of the second probability data and a preset formula for updating the second probability data, corresponding to the section, and associated with the first probability data;
s222, updating the first verification data according to a preset formula for updating the first verification data associated with the first probability data and the second probability data, adjusting the first probability data according to the position of the updated first verification data in a preset value range, and executing S221 and S222 in a circulating manner until the first verification data is in a corresponding preset range;
s223, updating the second check data according to a preset formula for updating the second check data, wherein the preset formula is associated with the first probability data, the second probability data and the first check data, adjusting the first probability data according to the position of the updated second check data in a preset value range, and executing S221, S222 and S223 in a circulating mode until the second check data is in the corresponding preset range;
s224, updating the third check data according to a preset formula for updating the third check data associated with the first probability data, the second probability data, the first check data and the second check data, adjusting the first probability data according to a position of the updated third check data in a preset value range, and executing S221, S222, S223 and S224 in a loop until the third check data is in the preset range corresponding to the third probability data;
and S225, updating the fourth verification data according to a preset formula for updating the fourth verification data related to the first probability data, the second probability data, the first verification data, the second verification data and the third verification data, adjusting the first probability data according to the position of the updated fourth verification data in a preset value range, and executing S221, S222, S223, S224 and S225 in a circulating manner until the fourth verification data is in a preset range corresponding to the fourth verification data and the first probability data and the second probability data are in preset target ranges corresponding to the first probability data, the second probability data and the third verification data respectively.
The beneficial effect of adopting the further scheme is that: the values of the probabilities representing the gender and the age of the user are continuously adjusted in a recursion mode to gradually approach the maximum probability, so that the accuracy of judging the gender and the age of the user is greatly improved.
Further, the S3 further includes: and according to a content tag preset in a content database associated with the mobile equipment and the historical access record data of the user, obtaining content data matched with the characteristic data of the user.
The beneficial effect of adopting the further scheme is that: by combining the user characteristic data with the historical data of the user, the recommendation accuracy of the content data is improved greatly, and the receiving and using degree of the recommended content data by the user is greatly improved.
A content data recommendation system comprising:
the device comprises a raw data acquisition module, a data processing module and a data processing module, wherein the raw data acquisition module is used for acquiring physiological data of a user and motion data of shaking the movable device through a sensor on the movable device when the user uses the movable device;
the characteristic data acquisition module is used for processing the acquired physiological data and the acquired action data of the user to obtain the characteristic data of the user;
and the content recommendation module is used for obtaining content data matched with the characteristic data of the user according to a preset content tag in a content database associated with the mobile equipment and pushing the content data to the user.
The invention has the beneficial effects that: the method and the device can accurately push the appropriate content data to the user according to the real-time situation of the user, do not need to spend a large amount of time on collecting the historical data of the user, and improve the receiving and using degree of the pushed content data of the user.
Further, the physiological data in the raw data acquisition module is body temperature, and the motion data includes: the motion strength, frequency, speed, amplitude and direction; the characteristic data of the user comprises: sex and age.
The beneficial effect of adopting the further scheme is that: the gender and the age of the user are judged by collecting the body temperature and the action data of the user, the characteristic data of the user can be obtained in real time, and the content data of the composite user characteristic can be accurately pushed.
Further, the feature data acquisition module comprises the following sub-modules:
an initial data processing sub-module for obtaining first probability data representing a gender of the user, second probability data representing an age of the user, and a plurality of check data for correcting the first probability data and the second probability data according to the physiological data and the motion data;
a data adjustment and check sub-module, configured to adjust the first probability data according to the plurality of check data, and update the second probability data and the plurality of check data according to preset formulas associated with the first probability data and corresponding to the second probability data and the plurality of check data, so that the plurality of check data are all in preset ranges corresponding to the plurality of check data, and the first probability data and the second probability data are all in preset target ranges corresponding to the first probability data and the second probability data;
and the characteristic data judgment submodule is used for obtaining the characteristic data of the user according to the range of the first probability data and the second probability data.
The beneficial effect of adopting the further scheme is that: the probability that the user is different in gender and age is corrected through the range of the plurality of check data, errors caused by judging the gender and the age of the user according to single data are avoided, and the accuracy of judging the gender and the age of the user is greatly improved.
Further, in the initial data processing sub-module, the plurality of check data includes: the first check data, the second check data, the third check data and the fourth check data;
the data adjustment check submodule comprises the following units:
the second probability data updating unit is used for updating the second probability data according to a preset section of the second probability data and a preset formula for updating the second probability data, corresponding to the section, and associated with the first probability data;
the first check data updating unit is used for updating the first check data according to a preset formula for updating the first check data related to the first probability data and the second probability data, adjusting the first probability data according to the position of the updated first check data in a preset value range, and circularly calling the second probability data updating unit and the unit until the first check data is in the corresponding preset range;
the second check data updating unit is used for updating the second check data according to a preset formula for updating the second check data, which is associated with the first probability data, the second probability data and the first check data, adjusting the first probability data according to the position of the updated second check data in a preset value range, and circularly calling the second probability data updating unit, the first check data updating unit and the unit until the second check data is in the preset range corresponding to the second check data;
the third verification data updating unit is used for updating third verification data according to a preset formula for updating the third verification data, wherein the preset formula is associated with the first probability data, the second probability data, the first verification data and the second verification data;
and the data adjusting and terminating unit is used for updating the fourth verification data according to a preset formula for updating the fourth verification data associated with the first probability data, the second probability data, the first verification data, the second verification data and the third verification data, adjusting the first probability data according to the position of the updated fourth verification data in a preset value range, and circularly calling the second probability data updating unit, the first verification data updating unit, the second verification data updating unit, the third verification data updating unit and the unit until the fourth verification data is in a corresponding preset range and the first probability data and the second probability data are in corresponding preset target ranges respectively.
The beneficial effect of adopting the further scheme is that: the values of the probabilities representing the gender and the age of the user are continuously adjusted in a recursion mode to gradually approach the maximum probability, so that the accuracy of judging the gender and the age of the user is greatly improved.
Further, the content recommendation module is further configured to obtain content data matched with the feature data of the user according to a content tag preset in a content database associated with the mobile device in combination with historical access record data of the user.
The beneficial effect of adopting the further scheme is that: by combining the user characteristic data with the historical data of the user, the recommendation accuracy of the content data is improved greatly, and the receiving and using degree of the recommended content data by the user is greatly improved.
Drawings
Fig. 1 is a flowchart of a content data recommendation method in embodiment 1 of the present invention;
fig. 2 is a block diagram of a content data recommendation system in embodiment 2 of the present invention;
fig. 3 is a block diagram of the structure of each unit of a data adjustment check submodule in a content data recommendation system in embodiment 2 of the present invention.
Detailed Description
The principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.
Example 1
As shown in fig. 1, a content data recommendation method includes the steps of:
s1, when the user uses the mobile device, acquiring physiological data of the user and motion data of shaking the mobile device through a sensor on the mobile device;
s2, processing the acquired physiological data and the acquired action data of the user to obtain characteristic data of the user;
and S3, obtaining the content data matched with the characteristic data of the user according to the preset content label in the content database associated with the mobile device, and pushing the content data to the user.
Specifically, the mobile device may be various handheld devices, wearable devices, and the like, and the motion data of the user is acquired by shaking the mobile device by the user.
According to the method in the embodiment, the appropriate content data can be accurately pushed to the user according to the real-time condition of the user, a large amount of time is not needed to be spent on collecting the historical data of the user, and the receiving and using degree of the pushed content data by the user is improved.
Further, in S1, the physiological data is body temperature, and the motion data includes: the motion strength, frequency, speed, amplitude and direction; the characteristic data of the user includes: sex and age.
Specifically, the user can also use the handheld device sensor to acquire basic data strength, speed, frequency, amplitude, direction, temperature, sound, humidity and gravity of the current user using the handheld device through the related behavior action of the user using the handheld device. And then analyzing the data information captured by the sensor of the mobile equipment to obtain the characteristic data of the user, such as the gender, age group, mood, habit, character characteristic, environment and the like. And obtaining the user image through the characteristic data of the user, and accurately recommending the corresponding content data to the user for use.
For example, the method is used in a mobile phone app application for video, and the strength, speed, frequency, amplitude and plane direction of a user shaking the mobile phone are obtained by using a shaking-shaking function. The data are used for analyzing user portraits of user characteristic data such as gender, age bracket, mood, habit, character characteristic and the like of the user, and videos are recommended to the user to watch through the user portraits.
For example, the method is used in a mobile phone app application for selling commodities, and the user data such as the gender, age, mood, environment and regional characteristics of the user are analyzed by using the information of the movement speed, frequency, plane direction, temperature, sound, humidity and gravity provided by a mobile phone sensor, so that the user image at the time is generated by the user data, and the commodity data is recommended to the user.
When the method is used in mobile phone app application serving as a video, a plurality of labels can be added to all movie contents in a movie content database, for example, labels suitable for age groups are available: 0-7 years old, 7-15 years old, 15-23 years old, 23-35 years old, 35-50 years old, 50-60 years old, 60-70 years old, and over 70 years old, etc.; type label: wars, science fiction, inspirations, ethics, families, fashion, speech, animation, singing and dancing, records, disasters and other types of labels. The same movie content can correspond to various labels, and can be in a one-to-many relationship, after the user characteristics are calculated through the big data, the same movie content is matched with the labels of the movie content in the movie content database through the user characteristics, and the movie content with the most matched label content is watched by the recommended user. And if a plurality of contents are matched, sorting with the highest click rate and pushing in a priority mode.
The method can also be used for various scenes such as a video screen pushing and playing function by shaking, an audio recommending and playing function by shaking, an E-commerce commodity recommending and selling function by shaking, a merchant activity recommending function by shaking, a recommending application by shaking, downloading and installing and the like.
In the improvement, the gender and the age of the user are judged by collecting the body temperature and the action data of the user, the characteristic data of the user can be obtained in real time, and the content data of the composite user characteristic can be accurately pushed.
Further, in S2, the process of processing to obtain the feature data of the user is:
s21, obtaining an initial value of first probability data indicating a gender of the user, an initial value of second probability data indicating an age of the user, and initial values of a plurality of pieces of check data for correcting the first probability data and the second probability data, based on the physiological data and the motion data;
s22, adjusting the first probability data according to the plurality of check data, updating the second probability data and the plurality of check data according to preset formulas which are respectively corresponding to the second probability data and the plurality of check data and are associated with the first probability data, and enabling the plurality of check data to be in preset ranges respectively corresponding to the second probability data and the plurality of check data to be in preset target ranges respectively corresponding to the first probability data and the second probability data;
and S23, obtaining the characteristic data of the user according to the range of the values of the first probability data and the second probability data.
Specifically, a represents the action strength, and the unit is as follows: n (cattle), e.g. 1N, b represents the amplitude of the movement in M (meters), e.g. 0.2M, and there are certain differences in normal height between people of different ages, typically this value is between 0.01 and 0.5, c represents the angle of movement, e.g.: 30 °, d represents the frequency of motion, in units of time/second, as: 1/sec, e represents body temperature, in degrees c, such as: 37 ℃ is carried out.
According to medical statistics, generally speaking, women have an average body temperature 0.3 ℃ higher than men. And the difference in human body temperature is 0-1.5 ℃ at different time periods. Through analysis of the time period and the reported body temperature e, first probability data A representing the probability of the gender of the user is obtained according to an average distribution method, the value of the data A represents the probability that the user is male, and if A is 0.4, the probability that the user has 40% probability of male and 60% probability of female is represented.
The initial value of the second probability data B is calculated by calculating the formula a x B x d, the initial value of the data B generally ranges from 0 to 10, and the value of the data B is used to represent the probability that the user is in each age group.
The initial value of the first verification data C is obtained by calculation through the formula a × b × d/cos (C), and the initial value of the data C generally floats between 0 and 100.
The initial value of the second check data D is calculated by calculating the formula a × b × sin (c), and the initial value of the data D generally floats between 0 and 5.
The initial value of the third verification data E is obtained by calculating the formula a/b sin (c), and the initial value of the data E generally floats between 0 and 200.
The initial value of the fourth verification data F is obtained by calculating formula b x d, and the initial value of the data F generally floats between 0 and 5.
The values of B, C, D, E, F are adjusted to the preset range by actively adjusting the value of the first probability data A, and the value of A is also adjusted to the preset target range, so as to obtain the gender and age of the user.
In the improvement, the probability that the user is different in gender and age is corrected through the range of the plurality of check data, errors caused by judging the gender and the age of the user according to single data are avoided, and the accuracy of judging the gender and the age of the user is greatly improved.
Further, in S21, the plurality of check data includes: the first check data, the second check data, the third check data and the fourth check data; s22 further includes:
s221, updating second probability data according to a preset section of the second probability data and a preset formula for updating the second probability data, corresponding to the section, and associated with the first probability data;
s222, updating the first verification data according to a preset formula for updating the first verification data associated with the first probability data and the second probability data, adjusting the first probability data according to the position of the updated first verification data in a preset value range, and executing S221 and S222 in a circulating manner until the first verification data is in a corresponding preset range;
s223, updating the second check data according to a preset formula for updating the second check data, wherein the preset formula is associated with the first probability data, the second probability data and the first check data, adjusting the first probability data according to the position of the updated second check data in a preset value range, and executing S221, S222 and S223 in a circulating mode until the second check data is in the corresponding preset range;
s224, updating the third check data according to a preset formula for updating the third check data associated with the first probability data, the second probability data, the first check data and the second check data, adjusting the first probability data according to a position of the updated third check data in a preset value range, and executing S221, S222, S223 and S224 in a loop until the third check data is in the preset range corresponding to the third probability data;
and S225, updating the fourth verification data according to a preset formula for updating the fourth verification data related to the first probability data, the second probability data, the first verification data, the second verification data and the third verification data, adjusting the first probability data according to the position of the updated fourth verification data in a preset value range, and executing S221, S222, S223, S224 and S225 in a circulating manner until the fourth verification data is in the preset range corresponding to the fourth verification data and the first probability data and the second probability data are in the preset target ranges corresponding to the first probability data, the second probability data and the third verification data respectively.
Specifically, first, the value range of B is divided into several sub-range segments according to the initial value of the first probability data B, for example, the value range is described as being divided into 5 segments. The subrange Bq1 is 0 to 2, meaning an age of 2 to 5 years or over 70 years; the sub-range Bq2 is 2-4, indicating an age range of 5-15 years; the sub-range Bq3 is 4-6, indicating an age range of 55-70 years; the subrange Bq4 from 4 to 8, indicating an age group of 15 to 20 years or 45 to 55 years; the self-range Bq5 is 8-10, indicating an age range of 20-45 years. The first round of calculation is performed based on the initial value of B in combination with the value of a, when the value of B is within the range of Bq1, the calculation formula of the sub-value B1 of B corresponding to Bq1 is B1 ═ B/2, and the calculation formula of the sub-value B2 of B corresponding to Bq2 is B2 ═ B/4, and by this rule, the values of the five sub-values B1-B5 of B corresponding to Bq1 to Bq5 are calculated, and the subsequent calculation is described by taking B1 as an example only, while in the actual calculation, calculation is performed with respect to the values of B1-B5.
Updating the value of C by combining the range value of A with the range value of B1, such as when A is 0.5 and B1 is 0.8, if the initial value of C is 20, then updating the value of C to C B12A (the formula is a preset formula), and the updated value of C is 6.4. If the updated value of C is less than 5, the value does not belong to the preset range, the value is judged to be abnormal, the value of A needs to be increased by 0.005 for operation, and if the updated value of C is greater than the value of C before updating2If the value is not within the preset range, the value of A is also abnormal, the value of A needs to be reduced by 0.005, and then the above process is repeated to recheck the updated values of B1-B5 and C until the updated value of C is within the preset range.
If the updated value of C falls within the preset range, it is judged as normal, and the calculation by the calculation formula with respect to the second calibration data D is performed: d3A B1/C (the formula is a preset formula) is calculated to obtain an updated D value, if the updated D value is smaller than the D/8 before updating or larger than 0.8D before updating, the D value does not belong to a preset range, the D value is judged to be abnormal, the two branch directions of increasing and decreasing the A value by 0.005 are respectively carried out for the values larger and smaller than the A value, then the above process is repeated, and the updated B1-B5, the updated C and the updated D value are checked again until the updated C and the updated D value belong to the preset range.
And if the updated value of D belongs to the preset range, judging that the D is normal, calculating the value of the updated E by calculating the value of a formula E (A + B1/C + D) (the formula is a preset formula) relative to the third check data E, if the value of the updated E is less than 1 and more than 20, judging that the E is abnormal, increasing and decreasing the value of the A in two branch directions by 0.005, repeating the above process, and rechecking the values of the updated B1-B5, the updated C, the updated D and the updated E until the values of the updated C, the updated D and the updated E all belong to the preset range.
And if the updated value of E belongs to the preset range, judging that the E is normal, calculating to obtain the updated value of F through an operation formula F of fourth check data F, wherein the operation formula F is F/E4 (the formula is a preset formula), and the updated value of F is between 0 and 0.5, and is the preset range, otherwise, the value of A needs to be increased and decreased by 0.005 in two branch directions for operation, and then repeating the process, and re-checking the values of B1-B5, updated C, updated D, updated E and updated F until the updated values of C, updated D, updated E and updated F all belong to the preset range. And (4) circularly adjusting, judging and calculating through recursive logic processing until the adjusted first probability data A reaches the target range of 0-0.1 or 0.9-1 and one item in B1-B5 reaches the target range of more than 0.9, namely the probability of one block in Bq1-Bq5 reaches more than 0.9 first. According to the gender and the age of the user, the gender and the age of the user can be obtained.
In the improvement, the values of the probabilities representing the gender and the age of the user are continuously adjusted in a recursion mode to gradually approach the maximum probability, so that the accuracy of judging the gender and the age of the user is greatly improved.
Further, S3 further includes: and according to a content tag preset in a content database associated with the mobile equipment, combining with the historical access record data of the user, obtaining content data matched with the characteristic data of the user.
Specifically, if the history of the user can be found in the database, the history data of the user and the feature data of the current user can be combined together by the colleague to match with the tag, so that the content data with the most matched tag content can be recommended to the user. And if a plurality of contents are matched, sorting with the highest click rate and pushing in a priority mode.
In the improvement, the accuracy of recommending the content data is improved greatly by combining the user characteristic data with the historical data of the user, so that the receiving and using degree of the recommended content data by the user is greatly improved.
Example 2
As shown in fig. 2, a content data recommendation system includes:
the device comprises a raw data acquisition module, a data processing module and a data processing module, wherein the raw data acquisition module is used for acquiring physiological data of a user and motion data of shaking the movable device through a sensor on the movable device when the user uses the movable device;
the characteristic data acquisition module is used for processing the acquired physiological data and the acquired action data of the user to obtain the characteristic data of the user;
and the content recommendation module is used for acquiring content data matched with the characteristic data of the user according to a preset content tag in a content database associated with the mobile equipment and pushing the content data to the user.
Specifically, the mobile device may be various handheld devices, wearable devices, and the like, and the motion data of the user is acquired by shaking the mobile device by the user.
The system in the embodiment can accurately push appropriate content data to the user according to the real-time condition of the user, does not need to spend a large amount of time on collecting historical data of the user, and improves the receiving and using degree of the pushed content data of the user.
Further, the physiological data in the original data acquisition module is body temperature, and the action data comprises: the motion strength, frequency, speed, amplitude and direction; the characteristic data of the user includes: sex and age.
Specifically, the user can also use the handheld device sensor to acquire basic data strength, speed, frequency, amplitude, direction, temperature, sound, humidity and gravity of the current user using the handheld device through the related behavior action of the user using the handheld device. And then analyzing the data information captured by the sensor of the mobile equipment to obtain the characteristic data of the user, such as the gender, age group, mood, habit, character characteristic, environment and the like. And obtaining the user image through the characteristic data of the user, and accurately recommending the corresponding content data to the user for use.
For example, the system is used in a mobile phone app application for video, and the strength, speed, frequency, amplitude and plane direction of a user shaking the mobile phone are obtained by using a shaking-shaking function. The data are used for analyzing user portraits of user characteristic data such as gender, age bracket, mood, habit, character characteristic and the like of the user, and videos are recommended to the user to watch through the user portraits.
For example, the system is used in a mobile phone app application for selling commodities, and the system analyzes user data such as the sex, age group, mood, environment and regional characteristics of a user by using the information of the moving speed, frequency, plane direction, temperature, sound, humidity and gravity provided by a mobile phone sensor, generates a user figure at that time by using the user data, and recommends commodity data for the user.
When the system is used in mobile phone app application serving as a video, a plurality of labels can be added to all movie contents in a movie content database, for example, a label suitable for age group: 0-7 years old, 7-15 years old, 15-23 years old, 23-35 years old, 35-50 years old, 50-60 years old, 60-70 years old, and over 70 years old, etc.; type label: wars, science fiction, inspirations, ethics, families, fashion, speech, animation, singing and dancing, records, disasters and other types of labels. The same movie content can correspond to various labels, and can be in a one-to-many relationship, after the user characteristics are calculated through the big data, the same movie content is matched with the labels of the movie content in the movie content database through the user characteristics, and the movie content with the most matched label content is watched by the recommended user. And if a plurality of contents are matched, sorting with the highest click rate and pushing in a priority mode.
The system can also be used for various scenes such as a video screen pushing and playing function by shaking, an audio recommending and playing function by shaking, an E-commerce commodity recommending and selling function by shaking, a merchant activity recommending function by shaking, a recommending application by shaking, downloading and installing and the like.
In the improvement, the gender and the age of the user are judged by collecting the body temperature and the action data of the user, the characteristic data of the user can be obtained in real time, and the content data of the composite user characteristic can be accurately pushed.
Further, the characteristic data acquisition module comprises the following sub-modules:
an initial data processing sub-module for obtaining first probability data representing a gender of the user, second probability data representing an age of the user, and a plurality of check data for correcting the first probability data and the second probability data, based on the physiological data and the motion data;
the data adjustment and verification submodule is used for adjusting the first probability data according to the plurality of verification data, updating the second probability data and the plurality of verification data according to preset formulas which are respectively corresponding to the second probability data and the plurality of verification data and are associated with the first probability data, and enabling the plurality of verification data to be in preset ranges respectively corresponding to the plurality of verification data and the first probability data and the second probability data to be in preset target ranges respectively corresponding to the first probability data and the second probability data;
and the characteristic data judgment submodule is used for obtaining the characteristic data of the user according to the range of the first probability data and the second probability data.
Specifically, a represents the action strength, and the unit is as follows: n (cattle), e.g. 1N, b represents the amplitude of the movement in M (meters), e.g. 0.2M, and there are certain differences in normal height between people of different ages, typically this value is between 0.01 and 0.5, c represents the angle of movement, e.g.: 30 °, d represents the frequency of motion, in units of time/second, as: 1/sec, e represents body temperature, in degrees c, such as: 37 ℃ is carried out.
According to medical statistics, generally speaking, women have an average body temperature 0.3 ℃ higher than men. And the difference in human body temperature is 0-1.5 ℃ at different time periods. Through analysis of the time period and the reported body temperature e, first probability data A representing the probability of the gender of the user is obtained according to an average distribution method, the value of the data A represents the probability that the user is male, and if A is 0.4, the probability that the user has 40% probability of male and 60% probability of female is represented.
The initial value of the second probability data B is calculated by calculating the formula a x B x d, the initial value of the data B generally ranges from 0 to 10, and the value of the data B is used to represent the probability that the user is in each age group.
The initial value of the first verification data C is obtained by calculation through the formula a × b × d/cos (C), and the initial value of the data C generally floats between 0 and 100.
The initial value of the second check data D is calculated by calculating the formula a × b × sin (c), and the initial value of the data D generally floats between 0 and 5.
The initial value of the third verification data E is obtained by calculating the formula a/b sin (c), and the initial value of the data E generally floats between 0 and 200.
The initial value of the fourth verification data F is obtained by calculating formula b x d, and the initial value of the data F generally floats between 0 and 5.
The values of B, C, D, E, F are adjusted to the preset range by actively adjusting the value of the first probability data A, and the value of A is also adjusted to the preset target range, so as to obtain the gender and age of the user.
In the improvement, the probability that the user is different in gender and age is corrected through the range of the plurality of check data, errors caused by judging the gender and the age of the user according to single data are avoided, and the accuracy of judging the gender and the age of the user is greatly improved.
Further, as shown in fig. 3, in the initial data processing sub-module, the plurality of check data includes: the first check data, the second check data, the third check data and the fourth check data;
the data adjustment check submodule comprises the following units:
the second probability data updating unit is used for updating the second probability data according to a preset section of the second probability data and a preset formula which is associated with the first probability data and is used for updating the second probability data and corresponds to the section;
the first check data updating unit is used for updating the first check data according to a preset formula for updating the first check data related to the first probability data and the second probability data, adjusting the first probability data according to the position of the updated first check data in a preset value range, and circularly calling the second probability data updating unit and the unit until the first check data is in the corresponding preset range;
the second check data updating unit is used for updating the second check data according to a preset formula for updating the second check data, which is associated with the first probability data, the second probability data and the first check data, adjusting the first probability data according to the position of the updated second check data in a preset value range, and circularly calling the second probability data updating unit, the first check data updating unit and the unit until the second check data is in the preset range corresponding to the second check data;
the third verification data updating unit is used for updating third verification data according to a preset formula for updating the third verification data, wherein the preset formula is associated with the first probability data, the second probability data, the first verification data and the second verification data;
and the data adjusting and terminating unit is used for updating the fourth verification data according to a preset formula for updating the fourth verification data associated with the first probability data, the second probability data, the first verification data, the second verification data and the third verification data, adjusting the first probability data according to the position of the updated fourth verification data in a preset value range, and circularly calling the second probability data updating unit, the first verification data updating unit, the second verification data updating unit, the third verification data updating unit and the unit until the fourth verification data is in a preset range corresponding to the fourth probability data and the first probability data and the second probability data are in preset target ranges corresponding to the fourth probability data and the third verification data respectively.
Specifically, first, the value range of B is divided into several sub-range segments according to the initial value of the first probability data B, for example, the value range is described as being divided into 5 segments. The subrange Bq1 is 0 to 2, meaning an age of 2 to 5 years or over 70 years; the sub-range Bq2 is 2-4, indicating an age range of 5-15 years; the sub-range Bq3 is 4-6, indicating an age range of 55-70 years; the subrange Bq4 from 4 to 8, indicating an age group of 15 to 20 years or 45 to 55 years; the self-range Bq5 is 8-10, indicating an age range of 20-45 years. The first round of calculation is performed based on the initial value of B in combination with the value of a, when the value of B is within the range of Bq1, the calculation formula of the sub-value B1 of B corresponding to Bq1 is B1 ═ B/2, and the calculation formula of the sub-value B2 of B corresponding to Bq2 is B2 ═ B/4, and by this rule, the values of the five sub-values B1-B5 of B corresponding to Bq1 to Bq5 are calculated, and the subsequent calculation is described by taking B1 as an example only, while in the actual calculation, calculation is performed with respect to the values of B1-B5.
Updating the value of C by combining the range value of A with the range value of B1, such as when A is 0.5 and B1 is 0.8, if the initial value of C is 20, then updating the value of C to C B12A (the formula is a preset formula), and the updated value of C is 6.4. If the updated value of C is less than 5, the value does not belong to the preset range, the value is judged to be abnormal, the value of A needs to be increased by 0.005 for operation, and if the updated value of C is greater than the value of C before updating2If the value is not within the preset range, the value of A is also abnormal, the value of A needs to be reduced by 0.005, and then the above process is repeated to recheck the updated values of B1-B5 and C until the updated value of C is within the preset range.
If the updated value of C falls within the preset range, it is judged as normal, and the calculation by the calculation formula with respect to the second calibration data D is performed: d3A B1/C (the formula is a preset formula) is calculated to obtain an updated D value, if the updated D value is smaller than the D/8 before updating or larger than 0.8D before updating, the D value does not belong to a preset range, the D value is judged to be abnormal, the two branch directions of increasing and decreasing the A value by 0.005 are respectively carried out for the values larger and smaller than the A value, then the above process is repeated, and the updated B1-B5, the updated C and the updated D value are checked again until the updated C and the updated D value belong to the preset range.
And if the updated value of D belongs to the preset range, judging that the D is normal, calculating the value of the updated E by calculating the value of a formula E (A + B1/C + D) (the formula is a preset formula) relative to the third check data E, if the value of the updated E is less than 1 and more than 20, judging that the E is abnormal, increasing and decreasing the value of the A in two branch directions by 0.005, repeating the above process, and rechecking the values of the updated B1-B5, the updated C, the updated D and the updated E until the values of the updated C, the updated D and the updated E all belong to the preset range.
And if the updated value of E belongs to the preset range, judging that the E is normal, calculating to obtain the updated value of F through an operation formula F of fourth check data F, wherein the operation formula F is F/E4 (the formula is a preset formula), and the updated value of F is between 0 and 0.5, and is the preset range, otherwise, the value of A needs to be increased and decreased by 0.005 in two branch directions for operation, and then repeating the process, and re-checking the values of B1-B5, updated C, updated D, updated E and updated F until the updated values of C, updated D, updated E and updated F all belong to the preset range. And (4) circularly adjusting, judging and calculating through recursive logic processing until the adjusted first probability data A reaches the target range of 0-0.1 or 0.9-1 and one item in B1-B5 reaches the target range of more than 0.9, namely the probability of one block in Bq1-Bq5 reaches more than 0.9 first. According to the gender and the age of the user, the gender and the age of the user can be obtained.
In the improvement, the values of the probabilities representing the gender and the age of the user are continuously adjusted in a recursion mode to gradually approach the maximum probability, so that the accuracy of judging the gender and the age of the user is greatly improved.
Further, the content recommending module is used for obtaining content data matched with the characteristic data of the user according to a content label preset in a content database associated with the mobile device and by combining historical access record data of the user.
Specifically, if the history of the user can be found in the database, the history data of the user and the feature data of the current user can be combined together by the colleague to match with the tag, so that the content data with the most matched tag content can be recommended to the user. And if a plurality of contents are matched, sorting with the highest click rate and pushing in a priority mode.
In the improvement, the accuracy of recommending the content data is improved greatly by combining the user characteristic data with the historical data of the user, so that the receiving and using degree of the recommended content data by the user is greatly improved.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (8)
1. A content data recommendation method characterized by comprising the steps of:
s1, when the user uses the mobile device, acquiring physiological data of the user and motion data of shaking the mobile device through a sensor on the mobile device;
s2, processing the acquired physiological data and the acquired action data of the user to obtain characteristic data of the user;
s3, obtaining content data matched with the characteristic data of the user according to a content label preset in a content database associated with the mobile device, and pushing the content data to the user;
the feature data of the user in S1 includes: sex and age;
in S2, the process of processing to obtain the feature data of the user is:
s21, obtaining first probability data indicating gender of the user, second probability data indicating age of the user, and a plurality of check data for correcting the first probability data and the second probability data based on the physiological data and the motion data;
s22, adjusting the first probability data according to the plurality of check data, updating the second probability data and the plurality of check data according to a preset formula associated with the first probability data corresponding to the second probability data and the plurality of check data, respectively, so that the plurality of check data are all in a preset range corresponding to the second probability data and the plurality of check data are all in a preset target range corresponding to the first probability data and the second probability data;
and S23, obtaining the characteristic data of the user according to the range of the values of the first probability data and the second probability data.
2. The content data recommendation method according to claim 1, wherein the physiological data in S1 is body temperature, and the motion data includes: motion force, frequency, speed, amplitude and direction.
3. The content data recommendation method according to claim 1, wherein in S21, said plurality of check data includes: the first check data, the second check data, the third check data and the fourth check data; the S22 further includes:
s221, updating second probability data according to a preset section of the second probability data and a preset formula for updating the second probability data, corresponding to the section, and associated with the first probability data;
s222, updating the first verification data according to a preset formula for updating the first verification data associated with the first probability data and the second probability data, adjusting the first probability data according to the position of the updated first verification data in a preset value range, and executing S221 and S222 in a circulating manner until the first verification data is in a corresponding preset range;
s223, updating the second check data according to a preset formula for updating the second check data, wherein the preset formula is associated with the first probability data, the second probability data and the first check data, adjusting the first probability data according to the position of the updated second check data in a preset value range, and executing S221, S222 and S223 in a circulating mode until the second check data is in the corresponding preset range;
s224, updating the third check data according to a preset formula for updating the third check data associated with the first probability data, the second probability data, the first check data and the second check data, adjusting the first probability data according to a position of the updated third check data in a preset value range, and executing S221, S222, S223 and S224 in a loop until the third check data is in the preset range corresponding to the third probability data;
and S225, updating the fourth verification data according to a preset formula for updating the fourth verification data related to the first probability data, the second probability data, the first verification data, the second verification data and the third verification data, adjusting the first probability data according to the position of the updated fourth verification data in a preset value range, and executing S221, S222, S223, S224 and S225 in a circulating manner until the fourth verification data is in a preset range corresponding to the fourth verification data and the first probability data and the second probability data are in preset target ranges corresponding to the first probability data, the second probability data and the third verification data respectively.
4. The content data recommendation method according to any one of claims 1 to 3, wherein said S3 further comprises: and according to a content tag preset in a content database associated with the mobile equipment and the historical access record data of the user, obtaining content data matched with the characteristic data of the user.
5. A content data recommendation system, comprising:
the device comprises a raw data acquisition module, a data processing module and a data processing module, wherein the raw data acquisition module is used for acquiring physiological data of a user and motion data of shaking the movable device through a sensor on the movable device when the user uses the movable device;
the characteristic data acquisition module is used for processing the acquired physiological data and the acquired action data of the user to obtain the characteristic data of the user;
the content recommendation module is used for obtaining content data matched with the characteristic data of the user according to a content tag preset in a content database associated with the mobile equipment and pushing the content data to the user;
the characteristic data of the user comprises: sex and age;
the characteristic data acquisition module comprises the following sub-modules:
an initial data processing sub-module for obtaining first probability data representing a gender of the user, second probability data representing an age of the user, and a plurality of check data for correcting the first probability data and the second probability data according to the physiological data and the motion data;
a data adjustment and check sub-module, configured to adjust the first probability data according to the plurality of check data, and update the second probability data and the plurality of check data according to preset formulas associated with the first probability data and corresponding to the second probability data and the plurality of check data, so that the plurality of check data are all in preset ranges corresponding to the plurality of check data, and the first probability data and the second probability data are all in preset target ranges corresponding to the first probability data and the second probability data;
and the characteristic data judgment submodule is used for obtaining the characteristic data of the user according to the range of the first probability data and the second probability data.
6. The content data recommendation system according to claim 5, wherein the physiological data in the raw data acquisition module is body temperature, and the action data comprises: motion force, frequency, speed, amplitude and direction.
7. The content data recommendation system according to claim 5, wherein, in said initial data processing sub-module, said plurality of verification data comprises: the first check data, the second check data, the third check data and the fourth check data;
the data adjustment check submodule comprises the following units:
the second probability data updating unit is used for updating the second probability data according to a preset section of the second probability data and a preset formula for updating the second probability data, corresponding to the section, and associated with the first probability data;
the first check data updating unit is used for updating the first check data according to a preset formula for updating the first check data related to the first probability data and the second probability data, adjusting the first probability data according to the position of the updated first check data in a preset value range, and circularly calling the second probability data updating unit and the unit until the first check data is in the corresponding preset range;
the second check data updating unit is used for updating the second check data according to a preset formula for updating the second check data, which is associated with the first probability data, the second probability data and the first check data, adjusting the first probability data according to the position of the updated second check data in a preset value range, and circularly calling the second probability data updating unit, the first check data updating unit and the unit until the second check data is in the preset range corresponding to the second check data;
the third verification data updating unit is used for updating third verification data according to a preset formula for updating the third verification data, wherein the preset formula is associated with the first probability data, the second probability data, the first verification data and the second verification data;
and the data adjusting and terminating unit is used for updating the fourth verification data according to a preset formula for updating the fourth verification data associated with the first probability data, the second probability data, the first verification data, the second verification data and the third verification data, adjusting the first probability data according to the position of the updated fourth verification data in a preset value range, and circularly calling the second probability data updating unit, the first verification data updating unit, the second verification data updating unit, the third verification data updating unit and the unit until the fourth verification data is in a corresponding preset range and the first probability data and the second probability data are in corresponding preset target ranges respectively.
8. The content data recommendation system according to any one of claims 5 to 7, wherein the content recommendation module is further configured to obtain the content data matching the feature data of the user according to a content tag preset in a content database associated with the mobile device in combination with the historical access record data of the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710031491.XA CN106897369B (en) | 2017-01-17 | 2017-01-17 | Content data recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710031491.XA CN106897369B (en) | 2017-01-17 | 2017-01-17 | Content data recommendation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106897369A CN106897369A (en) | 2017-06-27 |
| CN106897369B true CN106897369B (en) | 2020-10-09 |
Family
ID=59198879
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710031491.XA Active CN106897369B (en) | 2017-01-17 | 2017-01-17 | Content data recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106897369B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109799733B (en) * | 2017-11-17 | 2021-05-11 | 广州奥睿智能科技有限公司 | Control method and control system of robot |
| CN110020155A (en) | 2017-12-06 | 2019-07-16 | 广东欧珀移动通信有限公司 | User's gender identification method and device |
| CN112818250B (en) * | 2018-03-07 | 2024-03-08 | 创新先进技术有限公司 | Content recommendation method, device, electronic equipment and system |
| CN110321474B (en) * | 2019-05-21 | 2022-01-11 | 北京奇艺世纪科技有限公司 | Recommendation method and device based on search terms, terminal equipment and storage medium |
| CN111310059B (en) * | 2020-04-01 | 2023-11-21 | 东软睿驰汽车技术（沈阳）有限公司 | User intention positioning method and device based on aggregated resources |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101437079A (en) * | 2008-12-31 | 2009-05-20 | 深圳华为通信技术有限公司 | Remission method for mobile terminal user emotion and mobile terminal |
| CN104811487A (en) * | 2015-04-10 | 2015-07-29 | 惠州Tcl移动通信有限公司 | Information push method based on mobile terminal and server and system thereof |
| CN104917896A (en) * | 2015-06-12 | 2015-09-16 | 努比亚技术有限公司 | Data pushing method and terminal equipment |
| CN105022814A (en) * | 2015-07-08 | 2015-11-04 | 广东欧珀移动通信有限公司 | Information recommendation method and user terminal |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8964127B2 (en) * | 2012-07-27 | 2015-02-24 | TCL Research America Inc. | User-sensing remote control system and method |
| WO2014057486A1 (en) * | 2012-10-11 | 2014-04-17 | Sela Itai | System and method for providing content in real-time |
-
2017
- 2017-01-17 CN CN201710031491.XA patent/CN106897369B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101437079A (en) * | 2008-12-31 | 2009-05-20 | 深圳华为通信技术有限公司 | Remission method for mobile terminal user emotion and mobile terminal |
| CN104811487A (en) * | 2015-04-10 | 2015-07-29 | 惠州Tcl移动通信有限公司 | Information push method based on mobile terminal and server and system thereof |
| CN104917896A (en) * | 2015-06-12 | 2015-09-16 | 努比亚技术有限公司 | Data pushing method and terminal equipment |
| CN105022814A (en) * | 2015-07-08 | 2015-11-04 | 广东欧珀移动通信有限公司 | Information recommendation method and user terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106897369A (en) | 2017-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106897369B (en) | Content data recommendation method and system | |
| CN109543111B (en) | Recommendation information screening method and device, storage medium and server | |
| US20190379629A1 (en) | Methods and apparatus to estimate demographics of users employing social media | |
| US11935079B2 (en) | Conducting digital surveys that collect and convert biometric data into survey respondent characteristics | |
| CN109416820B (en) | Method and system for automatically determining and responding to user satisfaction | |
| US11620678B2 (en) | Advertising method, device and system, and computer-readable storage medium | |
| US11093781B2 (en) | Customized action based on video item events | |
| CN109446419B (en) | Method and device for recommending video | |
| JP2022003523A (en) | Methods and apparatus for compensating impression data for misattribution and/or non-coverage by database proprietor | |
| US20200134295A1 (en) | Electronic display viewing verification | |
| US20170206328A1 (en) | Healthcare server, healthcare server control method, and non-transitory computer readable medium | |
| CN105874423A (en) | Methods and apparatus to detect engagement with media presented on wearable media devices | |
| US20170208021A1 (en) | Adaptive nudge messages to motivate individuals to achieve certain wellness goals | |
| CN109543132A (en) | Content recommendation method, device, electronic equipment and storage medium | |
| CN105590240A (en) | Discrete calculating method of brand advertisement effect optimization | |
| US10740774B2 (en) | Reducing processing requirements to correct for bias in ratings data having interdependencies among demographic statistics | |
| CN101615223A (en) | Health data integrated system and method thereof | |
| US20190065692A1 (en) | Method and System for Determining Individualized Heath Scores | |
| WO2023125460A1 (en) | Sorting model training method and sorting method | |
| CN107767874A (en) | A kind of baby crying sound identification reminding method and system | |
| CN111465949A (en) | Information processing apparatus, information processing method, and program | |
| Gershuny et al. | CAPTURE24: Testing self-report time-use diaries against objective instruments in real time | |
| CN114245185B (en) | Video recommendation method, model training method, device, electronic equipment and medium | |
| CN111341416B (en) | Psychological stress assessment model processing method and related equipment | |
| US20210257064A1 (en) | Method, apparatus, electronic device and interactive method for correcting measured data, computer readable medium and measuring device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |