CN110659412A - Method and apparatus for providing personalized service in electronic device - Google Patents
Method and apparatus for providing personalized service in electronic device Download PDFInfo
- Publication number
- CN110659412A CN110659412A CN201910813555.0A CN201910813555A CN110659412A CN 110659412 A CN110659412 A CN 110659412A CN 201910813555 A CN201910813555 A CN 201910813555A CN 110659412 A CN110659412 A CN 110659412A
- Authority
- CN
- China
- Prior art keywords
- user
- user behavior
- feature vector
- model
- classification
- 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
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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- User Interface Of Digital Computer (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A method and apparatus for providing personalized services in an electronic device are provided, the method comprising: (A) collecting user behavior data; (B) determining the identity of the user according to the user behavior data; (C) and providing personalized services matched with the user identity to the user. According to the method and the device, different users can be distinguished according to the acquired user behavior data, so that personalized services are provided for the users in a targeted and accurate mode, and user experience is improved.
Description
Technical Field
The present invention relates generally to the field of artificial intelligence, and more particularly, to a method and apparatus for providing personalized services in an electronic device.
Background
With the development of information technology and internet, people gradually enter an information overload era from an information deficiency era, and a recommendation system is generated. The recommendation system can recommend interesting videos to the user and expand the viewing experience of the user.
Specifically, the recommendation method and the disadvantages of the existing recommendation system are as follows:
firstly, different users are identified through a face recognition technology, so as to recommend services (such as videos, messages and the like) interested by the identified users, however, not all electronic devices are provided with a camera for face recognition, and therefore, the scheme has no general applicability.
Secondly, the points of interest of the historical operation of each user in the divided time periods are counted based on the historical operation records of the users, so that the recommendations are made for the users, however, the points of interest of different users in the same time period are overlapped together in this way, and the accuracy of the recommendations and the user experience are not high.
It can be seen that the existing recommendation system cannot accurately provide corresponding recommendation services for different users.
Disclosure of Invention
An exemplary embodiment of the present invention is to provide a method for recommending a service and an electronic device, which can overcome a defect that an existing recommendation system cannot accurately provide corresponding recommendation services for different users.
According to an aspect of exemplary embodiments of the present invention, there is provided a method for providing a personalized service in an electronic device, including: (A) collecting user behavior data; (B) determining the identity of the user according to the user behavior data; (C) and providing personalized services matched with the user identity to the user.
Optionally, step (B) comprises: inputting the user behavior data into a pre-trained user classification model, and determining the user identity through the user classification model, wherein the user classification model is trained in the following way: collecting a user behavior data sample; extracting a user behavior feature vector from the user behavior data sample; inputting the user behavior feature vector into a deep learning algorithm model to obtain at least one user behavior feature vector model through the operation processing of the deep learning algorithm model; and carrying out classification evaluation on the at least one user behavior feature vector model by using a classification algorithm to obtain a user classification model.
Optionally, the step of performing classification evaluation on the at least one user behavior feature vector model by using a classification algorithm includes: and distributing a corresponding weight ratio to each user behavior feature vector model, and performing classification algorithm operation based on the at least one user behavior feature vector model and the distributed weight ratio.
Optionally, the accuracy of the weight ratio assignment increases with the number of times of training the user classification model, wherein when there is a user behavior feature vector model related to a voiceprint, the weight ratio of the user behavior feature vector model related to the voiceprint is the largest weight ratio among the weight ratios of all the user behavior feature vector models.
Optionally, step (B) further comprises: and when the user identity is not determined by the user classification model, based on the criterion of minimizing the variance of the similar characteristic vector model, retraining the user classification model by redistributing the weight ratio of the user behavior characteristic vector model based on the user behavior data.
Optionally, step (C) further comprises: and setting output parameters matched with the user identity of the electronic equipment.
Optionally, the output parameters include at least one of: output volume, screen display brightness, size of screen display window, display font and personalized theme.
Optionally, the user behavior data comprises at least one of: user operation data, user sound data, user habit setting data and user historical watching data; the personalized service comprises at least one of: playing video, pushing messages, playing voice, displaying pictures and playing music.
According to another aspect of exemplary embodiments of the present invention, there is provided an apparatus for providing a personalized service in an electronic device, including: the acquisition unit is used for acquiring user behavior data; the processing unit is used for determining the identity of the user according to the user behavior data; and the output unit is used for providing personalized services matched with the user identity for the user.
Optionally, the processing unit inputs the user behavior data into a pre-trained user classification model, and determines the user identity through the user classification model, wherein the user classification model is trained by: collecting a user behavior data sample; extracting a user behavior feature vector from the user behavior data sample; inputting the user behavior feature vector into a deep learning algorithm model to obtain at least one user behavior feature vector model through the operation processing of the deep learning algorithm model; and carrying out classification evaluation on the at least one user behavior feature vector model by using a classification algorithm to obtain a user classification model.
Optionally, the processing of performing classification evaluation on the at least one user behavior feature vector model by using a classification algorithm includes: and distributing a corresponding weight ratio to each user behavior feature vector model, and performing classification algorithm operation based on the at least one user behavior feature vector model and the distributed weight ratio.
Optionally, the accuracy of the weight ratio assignment increases with the number of times of training the user classification model, wherein when there is a user behavior feature vector model related to a voiceprint, the weight ratio of the user behavior feature vector model related to the voiceprint is the largest weight ratio among the weight ratios of all the user behavior feature vector models.
Optionally, when the user classification model does not determine the user identity, the processing unit further retrains the user classification model by re-assigning a weight ratio of the user behavior feature vector model based on the user behavior data based on minimization of variance of the similar feature vector model.
Optionally, the output unit further sets output parameters of the electronic device matching the user identity.
Optionally, the output parameters include at least one of: output volume, screen display brightness, size of screen display window, display font and personalized theme.
Optionally, the user behavior data comprises at least one of: user operation data, user sound data, user habit setting data and user historical watching data; the personalized service comprises at least one of: playing video, pushing messages, playing voice, displaying pictures and playing music.
According to another aspect of exemplary embodiments of the present invention, there is provided an electronic apparatus, wherein the electronic apparatus includes: a processor; a memory storing a computer program which, when executed by the processor, implements a method for providing personalized services in an electronic device according to the invention.
According to another aspect of exemplary embodiments of the present invention, there is provided a computer-readable storage medium storing a computer program which, when executed by a processor, implements a method for providing personalized services in an electronic device according to the present invention.
According to the method and the device for providing the personalized service in the electronic device, different users can be distinguished according to the acquired user behavior data, so that the personalized service is provided for the users in a targeted and accurate manner, and the user experience is improved.
Additional aspects and/or advantages of the present general inventive concept will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the general inventive concept.
Drawings
The above and other objects of exemplary embodiments of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings which illustrate exemplary embodiments, wherein:
fig. 1 illustrates a flowchart of a method for providing personalized services in an electronic device according to an exemplary embodiment of the present invention;
fig. 2 illustrates an example for providing personalized services in an electronic device according to an exemplary embodiment of the present invention;
fig. 3 illustrates a block diagram of an apparatus for providing personalized services in an electronic device according to an exemplary embodiment of the present invention.
Detailed Description
Reference will now be made in detail to the exemplary embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the like elements throughout. The embodiments are described below in order to explain the present invention by referring to the figures.
Fig. 1 illustrates a flowchart of a method for providing a personalized service in an electronic device according to an exemplary embodiment of the present invention. Here, the method may be implemented by an apparatus for providing personalized services in an electronic device, or may be implemented entirely by a computer program, for example, the method may be performed by an application installed in the electronic device for providing personalized services, or by a functional program implemented in an operating system of the electronic device. By way of example, the electronic device may be an electronic device with an artificial intelligence operation function, such as a smart television, a tablet computer, a smart phone, and a multimedia player.
As shown in fig. 1, in step S100, user behavior data is collected.
Here, the user behavior data may be usage data of the user on the electronic device for constructing, training a machine learning model as a sample, or as input data for obtaining a classification result. As an example, the user behavior data may include at least one of: user operation data, user sound data, user habit setting data and user historical viewing data.
For example, the user operation data may include a key speed of the remote controller, a key frequency of the remote controller, a touch point area of the touch screen, a touch duration of the touch screen, and the like; the user voice data may include user voice data, user voiceprint data, etc.; the user habit setting data may include an input method of a user habit, a volume range of the user habit, a font specification of the user habit, a screen display brightness of the user habit, a start/stop time of the electronic device of the user habit, and the like; the user historical viewing data may include viewing periods, viewing content, web pages visited, etc. by the user at different periods. Here, it should be understood that the user behavior data described above is only an example, and may also be other user behavior data besides the user behavior data described above, and the present invention is not limited in any way herein.
In step S200, the user identity is determined according to the user behavior data.
Here, the user identity may be a distinction of users currently using the electronic device according to age and/or gender, for example, the user identity may be a child, a teenager, a young adult, a middle-aged adult, an old woman, a man, an old man, a middle-aged woman, etc. As an example, the user identity may be determined from user behavior data by way of machine learning (pre-building a suitable machine learning model).
Specifically, the user behavior data may be input into a pre-trained user classification model, and the user identity may be determined by the user classification model. As an example, a user classification model for determining the identity of a user is trained by:
firstly: and collecting user behavior data samples, wherein the user behavior data samples are user behavior data used for constructing and training a user classification model. For example, the user operation data, the user sound data, the user habit setting data, the user history viewing data, and the like as described above.
Then: and extracting a user behavior feature vector from the user behavior data sample. In particular, a user behavior feature vector having a calibrated value (e.g., a particular value or range) may be extracted from the user behavior data sample. For example, a voice content feature vector can be extracted from the collected user voice data; characteristic vectors such as frequency spectrum, tone and the like with a specific range can be extracted from the collected user voiceprint data; the key velocity feature vector with a specific velocity range can be extracted from the key velocity of the remote controller.
And then: the user behavior feature vector is input to the deep learning algorithm model to obtain at least one user behavior feature vector model through operation processing (such as convolution operation, pooling operation, feature operation and the like) of the deep learning algorithm model. For example, after inputting the user's spectrum feature vector, tone feature vector, key velocity feature vector, and key frequency feature vector into the deep learning algorithm model, the user behavior feature vector model related to the user's voice and operation behavior features, such as a child voice feature vector model and a child habit setting feature vector model, can be obtained after passing through the deep learning algorithm model. By way of example, the deep learning algorithm model may be a recurrent neural network, a deep neural network, a convolutional neural network, and the like.
And finally: and carrying out classification evaluation on the at least one user behavior feature vector model by using a classification algorithm to obtain a user classification model. Specifically, regarding the step of performing classification evaluation on the at least one user behavior feature vector model by using a classification algorithm, as an example, a corresponding weight ratio may be first assigned to each user behavior feature vector model, and then a classification algorithm operation may be performed based on each user behavior feature vector model and the assigned weight ratio thereof, so as to obtain a user classification model.
Specifically, in the initial use, because the number of times of use by the user is small, the obtained user classification model is not sound yet, and the user classification has errors, at this time, the device applying the method can allocate different weight ratios to different user behavior feature vector models according to the proportion from large to small of the user classification possibility based on the obtained user behavior feature vector model. As an example, the accuracy of the weight ratio assignment increases with the number of times of training the user classification model, wherein when there is a user behavior feature vector model related to a voiceprint, the weight ratio of the user behavior feature vector model related to the voiceprint is the largest weight ratio among the weight ratios of all the user behavior feature vector models. When there is no user behavior feature vector model related to the voiceprint, an appropriate weight ratio may be assigned to the current user behavior feature vector model according to the proportion of the user classification possibility from large to small.
For example, when there are a user behavior feature vector model a related to a voiceprint, a user behavior feature vector model B related to a remote controller key frequency, and a user behavior feature vector model C related to a viewing time, the weight ratio of the user behavior feature vector model a, the user behavior feature vector model B, and the user behavior feature vector model C may be set to a%, B%, and C%, where a% > B% > C%, and a% + B% + C ═ 1.
For another example, when there are only the user behavior feature vector B related to the remote controller key frequency and the user behavior feature vector C related to the viewing time, the weight ratio of the user behavior feature vector B and the user behavior feature vector C may be set to d% and e%, where d% > e%, and d% + e% + 1.
In addition, it should be understood that the weight ratio may also be set to d% < e% according to the user classification possibility, and the present invention is not limited thereto.
In addition, in step S200, when the user classification model does not determine the user identity, the user classification model is retrained by re-assigning the weight ratio of the user behavior feature vector model based on the user behavior data according to the minimization of the variance of the similar feature vector model.
For example, when the user classification model does not determine the user identity, the user classification model may be retrained by using the user behavior data as the user behavior data sample (as described above), and after at least one user behavior feature vector model is obtained through the operation processing of the deep learning algorithm model, a new user classification model may be obtained by performing a classification algorithm operation by assigning a weight ratio different from that of the previous time.
On the other hand, after the user identity is determined by using the trained user classification model, in step S300, a personalized service matching the user identity is provided to the user. Here, the personalized service may include at least one of: playing video, pushing messages, playing voice, displaying pictures and playing music.
Furthermore, as an example, in addition to providing the personalized service, output parameters of the electronic device matching the user identity may be set. Here, the output parameter includes at least one of: output volume, screen display brightness, size of screen display window, display font and personalized theme.
Fig. 2 illustrates an example of providing a personalized service in an electronic device according to an exemplary embodiment of the present invention.
Assuming that the electronic device is a smart television, after user behavior data (for example, the current viewing time of the user is 9-11 am, the key operation frequency is 800 plus 1200 ms/time, the key characteristics of the remote controller are irregular, the frequently pressed keys are a main menu key and an OK key, the habitual volume is 25-50 db, the sound frequency is according with the characteristics of boys, the historical high-frequency speech words are waning teams, super flying knight, etc., the speech content is children's sound, the sound characteristics are children's sound, and the speech characteristics during the viewing process are cheerful sound and happy sound) as shown in (a) in fig. 2 are collected, the user behavior data are input into a pre-trained user classification model, and after the classification of the user classification model, the identity of the user can be determined as boy (shown in (B) in fig. 2), and the age is about 3-10, the user may be recommended with a television program of interest to boys, such as a waning team, super flying man, etc., and accordingly, the personalized theme of the smart tv may be set to a child mode, a volume may be set to 25 db, a brightness of a screen may be adjusted to a range most suitable for a comfort level of eyes of a child, as shown in (C) of fig. 2, and a picture may be finally presented as shown in (D) of fig. 2.
According to the method for providing the personalized service in the electronic equipment, different users can be distinguished according to the acquired user behavior data, so that the personalized service is provided for the users in a targeted and accurate manner, and the user experience is improved.
Fig. 3 illustrates a block diagram of an apparatus for providing personalized services in an electronic device according to an exemplary embodiment of the present invention. As shown in fig. 3, an apparatus for providing a personalized service in an electronic device according to an exemplary embodiment of the present invention includes: an acquisition unit 100, a processing unit 200 and an output unit 300.
Specifically, the acquisition unit 100 acquires user behavior data.
Here, the user behavior data may be usage data of the user on the electronic device for constructing, training a machine learning model as a sample, or as input data for obtaining a classification result. As an example, the user behavior data may include at least one of: user operation data, user sound data, user habit setting data and user historical viewing data.
For example, the user operation data may include a key speed of the remote controller, a key frequency of the remote controller, a touch point area of the touch screen, a touch duration of the touch screen, and the like; the user voice data may include user voice data, user voiceprint data, etc.; the user habit setting data may include an input method of a user habit, a volume range of the user habit, a font specification of the user habit, a screen display brightness of the user habit, a start/stop time of the electronic device of the user habit, and the like; the user historical viewing data may include viewing periods, viewing content, web pages visited, etc. by the user at different periods. Here, it should be understood that the user behavior data described above is only an example, and may also be other user behavior data besides the user behavior data described above, and the present invention is not limited in any way herein.
The processing unit 200 determines the user identity from the user behavior data.
Here, the user identity may be a distinction of users currently using the electronic device according to age and/or gender, for example, the user identity may be a child, a teenager, a young adult, a middle-aged adult, an old woman, a man, an old man, a middle-aged woman, etc. As an example, the user identity may be determined from user behavior data by way of machine learning (pre-building a suitable machine learning model).
Specifically, the processing unit 200 may input the user behavior data into a pre-trained user classification model, and determine the user identity through the user classification model. As an example, a user classification model for determining the identity of a user is trained by:
firstly: and collecting user behavior data samples, wherein the user behavior data samples are user behavior data used for constructing and training a user classification model. For example, the user operation data, the user sound data, the user habit setting data, the user history viewing data, and the like as described above.
Then: and extracting a user behavior feature vector from the user behavior data sample. In particular, a user behavior feature vector having a calibrated value (e.g., a particular value or range) may be extracted from the user behavior data sample. For example, a voice content feature vector can be extracted from the collected user voice data; characteristic vectors such as frequency spectrum, tone and the like with a specific range can be extracted from the collected user voiceprint data; the key velocity feature vector with a specific velocity range can be extracted from the key velocity of the remote controller.
And then: the user behavior feature vector is input to the deep learning algorithm model to obtain at least one user behavior feature vector model through operation processing (such as convolution operation, pooling operation, feature operation and the like) of the deep learning algorithm model. For example, after inputting the user's spectrum feature vector, tone feature vector, key velocity feature vector, and key frequency feature vector into the deep learning algorithm model, the user behavior feature vector model related to the user's voice and operation behavior features, such as a child voice feature vector model and a child habit setting feature vector model, can be obtained after passing through the deep learning algorithm model. By way of example, the deep learning algorithm model may be a recurrent neural network, a deep neural network, a convolutional neural network, and the like.
And finally: and carrying out classification evaluation on the at least one user behavior feature vector model by using a classification algorithm to obtain a user classification model. Specifically, regarding the step of performing classification evaluation on the at least one user behavior feature vector model by using a classification algorithm, as an example, a corresponding weight ratio may be first assigned to each user behavior feature vector model, and then a classification algorithm operation may be performed based on each user behavior feature vector model and the assigned weight ratio thereof, so as to obtain a user classification model.
Specifically, in the initial use, because the number of times of use by the user is small, the obtained user classification model is not sound yet, and the user classification has errors, at this time, the device applying the method can allocate different weight ratios to different user behavior feature vector models according to the proportion from large to small of the user classification possibility based on the obtained user behavior feature vector model. As an example, the accuracy of the weight ratio assignment increases with the number of times of training the user classification model, wherein when there is a user behavior feature vector model related to a voiceprint, the weight ratio of the user behavior feature vector model related to the voiceprint is the largest weight ratio among the weight ratios of all the user behavior feature vector models. When there is no user behavior feature vector model related to the voiceprint, an appropriate weight ratio may be assigned to the current user behavior feature vector model according to the proportion of the user classification possibility from large to small.
In addition, when the user classification model does not determine the user identity, the processing unit 200 retrains the user classification model by re-assigning the weight ratio of the user behavior feature vector model based on the user behavior data according to the minimization of the variance of the similar feature vector model.
For example, when the user classification model does not determine the user identity, the processing unit 200 may retrain the user classification model with the user behavior data as a user behavior data sample (as described above), and after obtaining at least one user behavior feature vector model through the operation processing of the deep learning algorithm model, may assign a weight ratio different from the previous time to perform a classification algorithm operation to obtain a new user classification model.
On the other hand, after the user identity is determined by using the trained user classification model, the output unit 300 provides the personalized service matched with the user identity to the user. Here, the personalized service may include at least one of: playing video, pushing messages, playing voice, displaying pictures and playing music.
Further, as an example, the output unit 300 may set an output parameter of the electronic device matching the user identity in addition to providing the personalized service. Here, the output parameter includes at least one of: output volume, screen display brightness, size of screen display window, display font and personalized theme.
Further, it should be understood that respective units in the apparatus for providing personalized services in an electronic device according to the exemplary embodiments of the present invention may be implemented as hardware components and/or software components. The individual units may be implemented, for example, using Field Programmable Gate Arrays (FPGAs) or Application Specific Integrated Circuits (ASICs), depending on the processing performed by the individual units as defined by the skilled person.
The computer-readable storage medium according to an exemplary embodiment of the present invention stores a computer program that, when executed by a processor, causes the processor to perform the method for providing a personalized service in an electronic device of the above-described exemplary embodiment. The computer readable storage medium is any data storage device that can store data which can be read by a computer system. Examples of computer-readable storage media include: read-only memory, random access memory, read-only optical disks, magnetic tapes, floppy disks, optical data storage devices, and carrier waves (such as data transmission through the internet via wired or wireless transmission paths).
An electronic device according to an exemplary embodiment of the present invention includes: a processor (not shown) and a memory (not shown), wherein the memory stores a computer program which, when executed by the processor, implements the method for providing personalized services in an electronic device as described in the above exemplary embodiments.
In summary, in the method and the device for providing personalized services in the electronic device according to the exemplary embodiments of the present invention, different users can be distinguished according to the obtained user behavior data, so that personalized services are provided to the users in a targeted and accurate manner, and user experience is improved.
While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention as defined by the following claims.
Claims (18)
1. A method for providing personalized services in an electronic device, comprising:
(A) collecting user behavior data;
(B) determining the identity of the user according to the user behavior data;
(C) and providing personalized services matched with the user identity to the user.
2. The method of claim 1, wherein step (B) comprises:
inputting the user behavior data into a pre-trained user classification model, determining the user identity through the user classification model,
wherein the user classification model is trained by:
collecting a user behavior data sample;
extracting a user behavior feature vector from the user behavior data sample;
inputting the user behavior feature vector into a deep learning algorithm model to obtain at least one user behavior feature vector model through the operation processing of the deep learning algorithm model;
and carrying out classification evaluation on the at least one user behavior feature vector model by using a classification algorithm to obtain a user classification model.
3. The method of claim 2, wherein the step of performing a classification evaluation on the at least one user behavior feature vector model using a classification algorithm comprises:
and distributing a corresponding weight ratio to each user behavior feature vector model, and performing classification algorithm operation based on the at least one user behavior feature vector model and the distributed weight ratio.
4. The method of claim 3, wherein the accuracy of the weight ratio assignment increases with an increasing number of times the user classification model is trained,
when the user behavior feature vector model related to the voiceprint exists, the weight ratio of the user behavior feature vector model related to the voiceprint is the largest weight ratio of the weight ratios of all the user behavior feature vector models.
5. The method of claim 3, wherein step (B) further comprises:
and when the user identity is not determined by the user classification model, based on the criterion of minimizing the variance of the similar characteristic vector model, retraining the user classification model by redistributing the weight ratio of the user behavior characteristic vector model based on the user behavior data.
6. The method of claim 1, wherein step (C) further comprises: and setting output parameters matched with the user identity of the electronic equipment.
7. The method of claim 6, wherein the output parameters comprise at least one of: output volume, screen display brightness, size of screen display window, display font and personalized theme.
8. The method of claim 1, wherein the user behavior data comprises at least one of: user operation data, user sound data, user habit setting data and user historical watching data; the personalized service comprises at least one of: playing video, pushing messages, playing voice, displaying pictures and playing music.
9. An apparatus for providing personalized services in an electronic device, comprising:
the acquisition unit is used for acquiring user behavior data;
the processing unit is used for determining the identity of the user according to the user behavior data;
and the output unit is used for providing personalized services matched with the user identity for the user.
10. The apparatus of claim 9, wherein the processing unit inputs the user behavior data into a pre-trained user classification model from which a user identity is determined,
wherein the user classification model is trained by:
collecting a user behavior data sample;
extracting a user behavior feature vector from the user behavior data sample;
inputting the user behavior feature vector into a deep learning algorithm model to obtain at least one user behavior feature vector model through the operation processing of the deep learning algorithm model;
and carrying out classification evaluation on the at least one user behavior feature vector model by using a classification algorithm to obtain a user classification model.
11. The apparatus of claim 10, wherein the process of performing a classification evaluation on the at least one user behavior feature vector model using a classification algorithm comprises:
and distributing a corresponding weight ratio to each user behavior feature vector model, and performing classification algorithm operation based on the at least one user behavior feature vector model and the distributed weight ratio.
12. The apparatus of claim 11, wherein an accuracy of the weight ratio assignment increases with an increasing number of times a user classification model is trained,
when the user behavior feature vector model related to the voiceprint exists, the weight ratio of the user behavior feature vector model related to the voiceprint is the largest weight ratio of the weight ratios of all the user behavior feature vector models.
13. The apparatus of claim 12, wherein the processing unit is further to retrain the user classification model by re-assigning a weight ratio of the user behavior feature vector model based on the user behavior data based on minimizing a variance of the similar feature vector model when the user classification model does not determine the user identity.
14. The apparatus of claim 9, wherein the output unit further sets output parameters of the electronic apparatus matching the user identity.
15. The apparatus of claim 14, wherein the output parameters comprise at least one of: output volume, screen display brightness, size of screen display window, display font and personalized theme.
16. The device of claim 9, wherein the user behavior data comprises at least one of: user operation data, user sound data, user habit setting data and user historical watching data; the personalized service comprises at least one of: playing video, pushing messages, playing voice, displaying pictures and playing music.
17. An electronic device, wherein the electronic device comprises:
a processor;
memory storing a computer program which, when executed by a processor, implements a method for providing personalized services in an electronic device as claimed in any one of claims 1 to 8.
18. A computer-readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the method for providing personalized services in an electronic device according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910813555.0A CN110659412A (en) | 2019-08-30 | 2019-08-30 | Method and apparatus for providing personalized service in electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910813555.0A CN110659412A (en) | 2019-08-30 | 2019-08-30 | Method and apparatus for providing personalized service in electronic device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110659412A true CN110659412A (en) | 2020-01-07 |
Family
ID=69036854
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910813555.0A Pending CN110659412A (en) | 2019-08-30 | 2019-08-30 | Method and apparatus for providing personalized service in electronic device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110659412A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111459784A (en) * | 2020-04-08 | 2020-07-28 | 广州三星通信技术研究有限公司 | Terminal operation processing method and device |
| CN111736940A (en) * | 2020-06-24 | 2020-10-02 | 中国银行股份有限公司 | Business interface display method and device for intelligent counter |
| CN112632369A (en) * | 2020-12-05 | 2021-04-09 | 武汉风行在线技术有限公司 | Short video recommendation system and method for identifying laughter |
| CN112785267A (en) * | 2021-01-23 | 2021-05-11 | 南京利特嘉软件科技有限公司 | Flight information management method and system based on MVC framework technology |
| CN113497980A (en) * | 2020-03-20 | 2021-10-12 | 佛山市云米电器科技有限公司 | Interface mode display method, cloud server, television, system and storage medium |
| CN113497979A (en) * | 2020-03-20 | 2021-10-12 | 佛山市云米电器科技有限公司 | Interface mode display method, cloud server, television, system and storage medium |
| CN114143607A (en) * | 2020-09-03 | 2022-03-04 | 惠州视维新技术有限公司 | Information recommendation method, storage medium and smart television |
| CN114598932A (en) * | 2020-12-04 | 2022-06-07 | 深圳Tcl新技术有限公司 | Program recommendation method, terminal and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106412638A (en) * | 2016-09-26 | 2017-02-15 | 天脉聚源（北京）科技有限公司 | Method and apparatus for determining user category |
| CN107944915A (en) * | 2017-11-21 | 2018-04-20 | 北京深极智能科技有限公司 | A kind of game user behavior analysis method and computer-readable recording medium |
| CN108076154A (en) * | 2017-12-21 | 2018-05-25 | 广东欧珀移动通信有限公司 | Application message recommends method, apparatus and storage medium and server |
| CN108628886A (en) * | 2017-03-20 | 2018-10-09 | 中移（杭州）信息技术有限公司 | A kind of audio file recommendation method and device |
| CN108737872A (en) * | 2018-06-08 | 2018-11-02 | 百度在线网络技术（北京）有限公司 | Method and apparatus for output information |
| CN109146076A (en) * | 2018-08-13 | 2019-01-04 | 东软集团股份有限公司 | model generating method and device, data processing method and device |
-
2019
- 2019-08-30 CN CN201910813555.0A patent/CN110659412A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106412638A (en) * | 2016-09-26 | 2017-02-15 | 天脉聚源（北京）科技有限公司 | Method and apparatus for determining user category |
| CN108628886A (en) * | 2017-03-20 | 2018-10-09 | 中移（杭州）信息技术有限公司 | A kind of audio file recommendation method and device |
| CN107944915A (en) * | 2017-11-21 | 2018-04-20 | 北京深极智能科技有限公司 | A kind of game user behavior analysis method and computer-readable recording medium |
| CN108076154A (en) * | 2017-12-21 | 2018-05-25 | 广东欧珀移动通信有限公司 | Application message recommends method, apparatus and storage medium and server |
| CN108737872A (en) * | 2018-06-08 | 2018-11-02 | 百度在线网络技术（北京）有限公司 | Method and apparatus for output information |
| CN109146076A (en) * | 2018-08-13 | 2019-01-04 | 东软集团股份有限公司 | model generating method and device, data processing method and device |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113497980A (en) * | 2020-03-20 | 2021-10-12 | 佛山市云米电器科技有限公司 | Interface mode display method, cloud server, television, system and storage medium |
| CN113497979A (en) * | 2020-03-20 | 2021-10-12 | 佛山市云米电器科技有限公司 | Interface mode display method, cloud server, television, system and storage medium |
| CN111459784A (en) * | 2020-04-08 | 2020-07-28 | 广州三星通信技术研究有限公司 | Terminal operation processing method and device |
| CN111736940A (en) * | 2020-06-24 | 2020-10-02 | 中国银行股份有限公司 | Business interface display method and device for intelligent counter |
| CN114143607A (en) * | 2020-09-03 | 2022-03-04 | 惠州视维新技术有限公司 | Information recommendation method, storage medium and smart television |
| CN114143607B (en) * | 2020-09-03 | 2024-03-15 | 惠州视维新技术有限公司 | Information recommendation method, storage medium and intelligent television |
| CN114598932A (en) * | 2020-12-04 | 2022-06-07 | 深圳Tcl新技术有限公司 | Program recommendation method, terminal and storage medium |
| CN112632369A (en) * | 2020-12-05 | 2021-04-09 | 武汉风行在线技术有限公司 | Short video recommendation system and method for identifying laughter |
| CN112785267A (en) * | 2021-01-23 | 2021-05-11 | 南京利特嘉软件科技有限公司 | Flight information management method and system based on MVC framework technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110659412A (en) | Method and apparatus for providing personalized service in electronic device | |
| US10819811B2 (en) | Accumulation of real-time crowd sourced data for inferring metadata about entities | |
| US11922934B2 (en) | Generating response in conversation | |
| CN106095833B (en) | Human-computer dialogue content processing method | |
| CN110035302B (en) | Information recommendation method and device, model training method and device, computing equipment and storage medium | |
| US8909636B2 (en) | Lifestyle collecting apparatus, user interface device, and lifestyle collecting method | |
| CN111258435B (en) | Comment method and device for multimedia resources, electronic equipment and storage medium | |
| CN109086439A (en) | Information recommendation method and device | |
| US10795560B2 (en) | System and method for detection and visualization of anomalous media events | |
| CN109478142B (en) | Methods, systems, and media for presenting a user interface customized for predicted user activity | |
| CN110222233B (en) | Video recommendation method and device, server and storage medium | |
| CN108573032A (en) | Video recommendation method and device | |
| CN103718166A (en) | Information processing apparatus, information processing method, and computer program product | |
| CN110366043A (en) | Method for processing video frequency, device, electronic equipment and readable medium | |
| CN109783656B (en) | Recommendation method and system of audio and video data, server and storage medium | |
| CN104021162A (en) | Method and device for grading multimedia resource | |
| CN108322317A (en) | A kind of account identification correlating method and server | |
| CN109739354A (en) | A kind of multimedia interaction method and device based on sound | |
| CN106991641A (en) | It is implanted into the method and device of picture | |
| CN110110204A (en) | A kind of information recommendation method, device and the device for information recommendation | |
| CN113111197A (en) | Multimedia content recommendation method, device, equipment and storage medium | |
| US20150254562A1 (en) | Two-model recommender | |
| US20150170035A1 (en) | Real time personalization and categorization of entities | |
| CN115228091A (en) | Game recommendation method, device, equipment and computer readable storage medium | |
| CN109726267B (en) | Story recommendation method and device for story machine |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200107 |