CN107391661B - Recommended word display method and device - Google Patents
Recommended word display method and device Download PDFInfo
- Publication number
- CN107391661B CN107391661B CN201710587081.3A CN201710587081A CN107391661B CN 107391661 B CN107391661 B CN 107391661B CN 201710587081 A CN201710587081 A CN 201710587081A CN 107391661 B CN107391661 B CN 107391661B
- Authority
- CN
- China
- Prior art keywords
- sub
- recommended
- pictures
- picture
- value
- 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
- G06F16/90—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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5838—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- User Interface Of Digital Computer (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a recommended word display method and device. The method comprises the following steps: when an interactive instruction triggered by a user is detected, M recommended words and the heat value of each recommended word are obtained, the M recommended words are determined according to historical interactive data of the user, a background picture of a current interactive interface is divided into N sub-pictures, the significant value of each sub-picture is determined, the significant value of each sub-picture is used for representing the attention degree of the sub-picture by the user, M sub-pictures with larger significant values are selected from the N sub-pictures as target sub-pictures according to the significant values of the N sub-pictures, the M recommended words are displayed in the M target sub-pictures according to the mapping relation between the preset heat value of the recommended word and the significant value of the target sub-pictures, the purpose that the display content is determined based on the personalized interactive data of the user is achieved, the accuracy of the recommended words is higher, the display position is determined based on the characteristics of the background picture and the heat of the recommended words and is more flexible, thus, user experience is improved.
Description
Technical Field
The invention relates to big data recommendation technology, in particular to a recommended word display method and device.
Background
With the continuous development of communication technology, the interaction between the user and the terminal device is more and more frequent. The user may interact with the terminal device by entering speech or text. In this process, in order to improve the user experience and the interaction efficiency, example words, which are called recommended words, may be displayed to the user before the user interacts.
Currently, the recommended word is an operator-determined fixed word. And when the recommended word is displayed, displaying the recommended word on a fixed position of an interactive interface of the terminal equipment.
However, in the above process, for different users, the fixed recommended word is displayed at a fixed position, the display content and the display mode are single, and the user experience is not high.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a recommended word display method and device.
In a first aspect, an embodiment of the present invention provides a method for displaying recommended words, including:
when an interactive instruction triggered by a user is detected, acquiring M recommended words and the heat value of each recommended word; the M recommended words are determined according to historical interaction data of the user, and M is an integer greater than or equal to 1;
dividing a background picture of a current interactive interface into N sub-pictures, and determining a significant value of each sub-picture; wherein the significance value of the sub-picture is used for representing the degree of attention of the sub-picture by a user, and N is an integer greater than or equal to M;
according to the sizes of the significant values of the N sub-pictures, M sub-pictures with larger significant values are selected from the N sub-pictures as target sub-pictures;
and displaying the M recommended words in the M target sub-pictures according to a mapping relation between the preset heat value of the recommended word and the significant value of the target sub-pictures.
In the method shown above, the displaying the M recommended words in the M target sub-pictures specifically includes:
displaying the M recommended words in the M target sub-pictures, wherein the larger the significance value of the target sub-picture is, the larger the heat value of the recommended words displayed in the target sub-picture is.
In the method shown above, the obtaining M recommended words and the heat value of each recommended word includes:
and acquiring the M recommended words and the heat value of each recommended word from a server.
In the method shown above, the obtaining M recommended words and the heat value of each recommended word includes:
determining a candidate recommended word set of the user according to the historical interaction data of the user, and determining the heat value of the candidate recommended word;
and selecting M candidate recommended words with larger heat values from the candidate recommended word set as the M recommended words according to the heat values of the candidate recommended words.
In the method shown above, the determining the heat value of the candidate recommended word includes:
according to the formula
Determining the heat value of each candidate recommended word; it is composed ofIn, x1Representing the number of occurrences, x, of the candidate recommended word in a first period2And representing the frequency of occurrence of the candidate recommended word in a second time period, wherein the second time period is closer to the current time than the first time period, and the current time is the time for acquiring the interactive instruction triggered by the user.
In the method shown above, before dividing the background picture of the current interactive interface into N sub-pictures, the method further includes:
determining a saliency map of the background picture and the sum of saliency values of all pixel points in the saliency map according to an Itti algorithm;
the dividing the background picture of the current interactive interface into N sub-pictures and determining the significance value of each sub-picture includes:
averagely dividing the background picture into P sub-pictures, and determining a sub-saliency map of each sub-picture according to the saliency map of the background picture; wherein P represents the number of sub-pictures in the length direction and width direction of the saliency map, (P-1)2＜M≤P2And P is an integer, P x P ═ N;
and determining the quotient of the sum of the significant values of all the pixel points in each sub-significant graph and the sum of the significant values of all the pixel points in the significant graph of the background picture as the significant value of each sub-picture.
In a second aspect, an embodiment of the present invention further provides a recommended word display device, including:
the acquisition module is used for acquiring M recommended words and the heat value of each recommended word when an interactive instruction triggered by a user is detected; the M recommended words are determined according to historical interaction data of the user, and M is an integer greater than or equal to 1;
the first determining module is used for dividing the background picture of the current interactive interface into N sub-pictures and determining the significant value of each sub-picture; wherein the significance value of the sub-picture is used for representing the degree of attention of the sub-picture by a user, and N is an integer greater than or equal to M;
a selection module, configured to select M sub-pictures with larger significance values from the N sub-pictures as target sub-pictures according to the significance values of the N sub-pictures;
and the display module is used for displaying the M recommended words in the M target sub-pictures according to the mapping relation between the preset heat value of the recommended word and the significant value of the target sub-picture.
In the above-described apparatus, the display module is specifically configured to:
displaying the M recommended words in the M target sub-pictures, wherein the larger the significance value of the target sub-picture is, the larger the heat value of the recommended words displayed in the target sub-picture is.
In the above-described apparatus, the obtaining module is specifically configured to:
and acquiring the M recommended words and the heat value of each recommended word from a server.
In the above-described apparatus, the obtaining module is specifically configured to:
determining a candidate recommended word set of the user according to the historical interaction data of the user, and determining the heat value of the candidate recommended word;
and selecting M candidate recommended words with larger heat values from the candidate recommended word set as the M recommended words according to the heat values of the candidate recommended words.
In the apparatus shown above, the determining, by the obtaining module, the heat value of the candidate recommended word includes:
according to the formula
Determining the heat value of each candidate recommended word; wherein x is1Representing the number of occurrences, x, of the candidate recommended word in a first period2And representing the frequency of occurrence of the candidate recommended word in a second time period, wherein the second time period is closer to the current time than the first time period, and the current time is the time for acquiring the interactive instruction triggered by the user.
In the apparatus as described above, the apparatus further comprises:
the second determination module is used for determining the saliency map of the background picture and the sum of the saliency values of all the pixel points in the saliency map according to the Itti algorithm;
the first determining module is specifically configured to:
averagely dividing the background picture into P sub-pictures, and determining a sub-saliency map of each sub-picture according to the saliency map of the background picture; wherein P represents the number of sub-pictures in the length direction and width direction of the saliency map, (P-1)2＜M≤P2And P is an integer, P x P ═ N;
and determining the quotient of the sum of the significant values of all the pixel points in each sub-significant graph and the sum of the significant values of all the pixel points in the significant graph of the background picture as the significant value of each sub-picture.
The method and the device for displaying recommended words provided by the embodiment of the invention acquire M recommended words and the heat value of each recommended word when an interactive instruction triggered by a user is detected, wherein the M recommended words are determined according to historical interactive data of the user, a background picture of a current interactive interface is divided into N sub-pictures, and the significance value of each sub-picture is determined, wherein the significance value of the sub-picture is used for indicating the degree of attention of the sub-picture by the user, M sub-pictures with larger significance values are selected from the N sub-pictures as target sub-pictures according to the significance values of the N sub-pictures, the M recommended words are displayed in the M target sub-pictures according to the mapping relation between the preset heat value of the recommended word and the significance value of the target sub-picture, on one hand, the determination of the recommended word and the heat value of each recommended word according to the historical interactive data of the user is realized, on the other hand, after the background picture is divided into a plurality of sub-pictures, M sub-pictures with high user attention degrees are selected as target sub-pictures, and finally M recommended words are displayed in the M target sub-pictures based on the mapping relation between the heat value and the significance value of the target sub-pictures. According to the method and the device for displaying the recommended words, different recommended words are determined for different users, each recommended word can be displayed in different areas of the background picture based on the characteristics of the background picture, the display content is determined based on the personalized interactive data of the users, the accuracy of the recommended words is higher, the display position is determined based on the characteristics of the background picture and the heat of the recommended words, the method and the device are more flexible, and therefore user experience is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic diagram of an application scenario of a recommended word display method according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a recommended word display method according to an embodiment of the present invention;
FIG. 3A is a diagram illustrating a background picture of the current interactive interface in the embodiment shown in FIG. 2;
FIG. 3B is a schematic diagram of the saliency map and the determined sub saliency map of the background picture shown in FIG. 3A;
FIG. 3C is a diagram illustrating M recommended words displayed in M target sub-pictures;
FIG. 4 is a diagram illustrating a variation of the user interface in the embodiment of FIG. 2;
fig. 5 is a schematic structural diagram of a recommended word display device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The recommended word display method provided by the embodiment of the invention can be executed by the terminal equipment. The terminal device in the embodiment of the invention can be a handheld device, a vehicle-mounted device, a wearable device, a computing device, a smart television, various forms of User Equipment (UE), a Mobile Station (MS), a terminal (terminal) and the like. Illustratively, the handheld device may be a smartphone, tablet, or the like.
Fig. 1 is a schematic diagram of an application scenario of a recommended word display method according to an embodiment of the present invention. The method for displaying the recommended word provided by the embodiment of the invention can be applied to a scene where a user interacts with terminal equipment. The interaction here may be a voice interaction or a text interaction. As shown in fig. 1, which illustrates a user interacting with a television set 11, either in voice or text. The terminal device may display the current interactive interface 12 when detecting the interactive instruction triggered by the user. In order to improve the user experience and the efficiency of the interaction, the terminal device may display the recommended word to the user on the current interaction interface 12. According to the recommendation word display method provided by the embodiment of the invention, on one hand, different recommendation words can be displayed for different users, and on the other hand, each recommendation word is displayed at different positions of a background picture for different background pictures of a current interactive interface. The final display effect is as shown in fig. 1, and each recommended word is displayed in an area of the background picture that is more concerned by the user.
The method for displaying recommended words provided by the embodiment of the invention comprises the steps of obtaining M recommended words and the heat value of each recommended word when an interactive instruction triggered by a user is detected, wherein the M recommended words are determined according to historical interactive data of the user, dividing a background picture of a current interactive interface into N sub-pictures, determining the significant value of each sub-picture, wherein the significant value of each sub-picture is used for representing the attention degree of the sub-picture by the user, selecting M sub-pictures with larger significant values from the N sub-pictures as target sub-pictures according to the significant values of the N sub-pictures, displaying the M recommended words in the M target sub-pictures according to the mapping relation between the preset heat value of the recommended word and the significant value of the target sub-pictures, on one hand, determining the recommended words and the heat value of each recommended word according to the historical interactive data of the user is realized, on the other hand, after the background picture is divided into a plurality of sub-pictures, M sub-pictures with high user attention degrees are selected as target sub-pictures, and finally M recommended words are displayed in the M target sub-pictures based on the mapping relation between the heat value and the significance value of the target sub-pictures. The recommended word display method provided by the embodiment of the invention determines different recommended words for different users, can display each recommended word in different areas of the background picture based on the characteristics of the background picture, has higher accuracy of the recommended word due to the fact that the display content is determined based on the personalized interactive data of the users, and is more flexible due to the fact that the display position is determined based on the characteristics of the background picture and the heat of the recommended word, thereby improving the user experience.
The technical solution of the present invention will be described in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.
Fig. 2 is a flowchart illustrating a recommended word display method according to an embodiment of the present invention. As shown in fig. 2, the method for displaying recommended words provided by the embodiment of the present invention includes the following steps:
s201: and when an interactive instruction triggered by a user is detected, acquiring M recommended words and the heat value of each recommended word.
The M recommended words are determined according to historical interaction data of the user, and M is an integer greater than or equal to 1.
Specifically, when the terminal device is a television, the user may trigger the interactive instruction in a variety of ways: the remote controller is used for triggering, the touch screen of the television is used for triggering, Application programs (APP) which are installed on other terminal devices and can control the television are used for triggering, and the voice interaction interface is used for triggering. When the terminal equipment is a mobile phone, a user can trigger through a touch screen or a voice interaction interface.
The interaction instruction in the embodiment of the present invention may be an instruction indicating that a user is about to perform voice interaction, or may be an instruction indicating that a user is about to perform text interaction. For example, if the user clicks an icon "microphone" representing voice interaction through the touch screen to trigger the voice interaction, the interaction instruction triggered by the user is an instruction indicating that the user is about to perform the voice interaction.
After the user performs some operations to trigger the interaction instruction, for example, the user triggers the interaction instruction through the touch screen, the processor of the terminal device may detect that the capacitance value of the corresponding position of the touch screen changes, and determine that the user triggers the interaction instruction.
The terminal equipment can acquire the M recommended words and the heat value of each recommended word through the following two implementation modes:
the first implementation mode comprises the following steps: and acquiring the M recommended words and the heat value of each recommended word from the server. The terminal device may send an acquisition request to the server to acquire the M recommended words and the heat value of each recommended word.
The size of M may be determined according to the size of the screen of the terminal device. In this implementation, the terminal device sends the historical interaction data of the user to the server. The server stores historical interaction data for the user. The historical interactive data in the embodiment of the invention refers to keywords which are input or spoken when a user interacts with the terminal equipment. The server may obtain historical interaction data of the user in a past period of time, for example, in the past 3 months, and select words with occurrence times greater than a preset threshold from the historical interaction data to form a candidate recommended word set. It is understood that the number of candidate recommended words in the candidate recommended word set is greater than or equal to M. Then, the server calculates the heat value of each candidate recommended word. The hot value of the candidate recommended word in the embodiment of the invention represents the degree of interest of the user in the candidate recommended word.
In one implementation, the server may be based on a formula
And determining the heat value of each candidate recommended word. Wherein x is1Representing the number of occurrences, x, of the candidate referral in the first period2Representing the number of occurrences of the candidate recommended word in a second time period, the second time period being closer to the current time than the first time period, the current time beingThe time of the user-triggered interactive instruction is fetched. The server may determine the current time from a timestamp of the acquisition request sent by the terminal device. For example, the first time period may be 2017.5.20-2017.5.22, the second time period may be 2017.5.23-2017.5.25, the current time is 2017.5.26, and "weather" appears 4 times in the first time period and 2 times in the second time period, the popularity value of the candidate referral "weather" is 0.5, and "drama" appears 10 times in the first time period and 12 times in the second time period, the popularity value of the candidate referral "drama" is 1.2.
In another implementation, the server may be based on a formula
And determining the heat value of each candidate recommended word. Wherein x is1And x2The meaning of the representation is the same as in the last implementation.
In yet another implementation, the server may be based on a formula
And determining the heat value of each candidate recommended word. Wherein x is1And x2The meaning of the representation is the same as in the two implementations described above.
After the server calculates the heat value of each candidate recommended word, according to the heat value, M candidate recommended words with larger heat values are selected from the candidate recommended word set to serve as M recommended words. One mode is that the server ranks the candidate recommended words according to the heat value of the candidate recommended words from large to small, and the top M candidate recommended words are selected as the M recommended words.
And after determining the M recommended words and the heat value of each recommended word, the server sends the M recommended words and the heat value of each recommended word to the terminal equipment.
The second implementation mode comprises the following steps: determining a candidate recommended word set of the user according to historical interaction data of the user, and determining a heat value of the candidate recommended word; and selecting M candidate recommended words with larger heat values from the candidate recommended word set as the M recommended words according to the heat values of the candidate recommended words.
In this implementation, the terminal device stores therein historical interaction data of the user. The terminal device may obtain historical interaction data of the user within a past period of time, for example, within past 3 weeks, select words whose occurrence times are greater than a preset threshold from the historical interaction data, and form a candidate recommended word set. It is understood that the number of candidate recommended words in the candidate recommended word set is greater than or equal to M. Then, the terminal device calculates the heat value of each candidate recommended word. The terminal equipment can be according to the formula
And determining the heat value of each candidate recommended word. The terminal equipment can also be according to the formula
Or a formula
And determining the heat value of each candidate recommended word. Wherein x is1Representing the number of occurrences, x, of the candidate referral in the first period2And representing the frequency of occurrence of the candidate recommended word in a second time period, wherein the second time period is closer to the current time than the first time period, and the current time is the time for acquiring the interactive instruction triggered by the user. After the terminal equipment calculates the heat value of each candidate recommended word, M candidate recommended words with larger heat values are selected from the candidate recommended word set as M recommended words according to the heat value. One mode is that the server ranks the candidate recommended words according to the heat value of the candidate recommended words from large to small, and the top M candidate recommended words are selected as the M recommended words.
The first implementation differs from the second implementation in that: in the first implementation mode, the terminal device does not need to store historical interactive data of the user, so that the storage resources of the terminal device are saved, and the terminal device does not need to perform the step of determining the recommended word and the heat value of the recommended word, so that the terminal device is suitable for the terminal device with low processing performance; in the second implementation manner, the terminal device does not need to interact with the server, so that on one hand, signaling is saved, and on the other hand, under the condition of poor network performance, the recommended word of the user and the popularity of the recommended word can still be determined according to the historical interaction data of the user.
Optionally, the heat value of each candidate recommended word may also be determined according to the service type to which the candidate recommended word belongs. At this time, the weight of each service type is considered in the heat value of each candidate recommended word. The second implementation manner is described as an example:
and after the terminal equipment determines the candidate recommended word set, determining the service type of each candidate recommended word. The terminal device may determine the service type to which each candidate recommended word belongs through semantic analysis, for example, analyze the semantics of the candidate recommended word "drama" to determine that it belongs to the movie service. The terminal device stores the mapping relation between each service type and the weight value thereof. The terminal equipment can be according to the formula
And determining the heat value of each recommended word. Wherein w represents a weight value of a service type to which the candidate recommended word belongs.
S202: and dividing the background picture of the current interactive interface into N sub-pictures, and determining the significance value of each sub-picture.
Wherein the saliency value of the sub-picture is used to represent the degree to which the sub-picture is focused by the user. N is an integer greater than or equal to M.
Specifically, the background picture of the current interactive interface in the embodiment of the present invention may be a picture selected from a picture library, or may be a picture of a last frame displayed on a display screen of the terminal device before the user triggers the interactive instruction.
It should be noted that the background picture of the current interactive interface may be displayed in full screen or only in a partial area of the screen of the television.
In the embodiment of the invention, the display position of the recommended word is determined by utilizing the image significance. The image significance is an important visual feature in the image, and represents the attention degree of human eyes to certain areas of the picture.
When the background picture of the current interactive interface is divided into N sub-pictures, the background picture may be divided into P × P sub-pictures on average. P represents the number of sub-pictures in the longitudinal direction and the width direction of the background picture. The P × P sub-pictures are equal in size. P × P ═ N. The fact that the sub-pictures are equal in size means that the number of pixel points included in the sub-pictures is equal.
More specifically, (P-1)2＜M≤P2And p is an integer. This means that the number of the divided sub-pictures needs to be as close as possible to the number of the recommended words, so as to achieve that the recommended words are uniformly distributed in the background picture as much as possible.
Embodiments of the present invention may determine the saliency value of each sub-picture based on the saliency map model proposed by Itti. The specific process is as follows: before S202, determining a saliency map of a background picture and the sum of saliency values of all pixel points in the saliency map according to an Itti algorithm; in the specific implementation process of S202, the background picture is divided into P × P sub-pictures on average, and the sub-saliency map of each sub-picture is determined according to the saliency map of the background picture; and determining the quotient of the sum of the significant values of all the pixel points in each sub-significant image and the sum of the significant values of all the pixel points in the significant image of the background image as the significant value of each sub-image.
When determining the saliency map of the background picture according to the Itti algorithm, the Itti algorithm may be called to determine the saliency map of the background picture. This is a mature algorithm and is not described in detail here. In the saliency map of the background picture, each pixel point corresponds to a saliency value, and the saliency value of the pixel point represents the degree of contrast between the pixel point and the surrounding background in the aspects of color, brightness, direction and the like. After the saliency map of the background picture is obtained, the sum of the saliency values of all the pixel points in the saliency map is determined.
The background picture is divided into P × P sub-pictures on average, which means that the saliency map of the background picture is also divided into P × P sub-saliency maps on average, and the sub-saliency map of each sub-picture can be determined according to the saliency map of the background picture. Optionally, the background picture may be rectangular, and correspondingly, the sub-picture is also rectangular. And normalizing the sum of the significant values of all the pixel points in each sub-significant image, namely determining the quotient of the sum of the significant values of all the pixel points in each sub-significant image and the sum of the significant values of all the pixel points in the significant image of the background image as the significant value of each sub-image. The significant value of each sub-picture is the normalized significant value of the sub-picture. It can be understood that, without normalization, the sum of the saliency values of all the pixel points in each sub-saliency map may be determined as the saliency value of the sub-picture. The larger the saliency value of a sub-picture is, the stronger the visual saliency of the sub-picture is.
S203: and according to the sizes of the significant values of the N sub-pictures, selecting M sub-pictures with larger significant values from the N sub-pictures as target sub-pictures.
Specifically, the significance values of the N sub-pictures may be sorted from large to small, and the top M sub-pictures with larger significance values are selected as target sub-pictures.
The target sub-picture in the embodiment of the present invention refers to a sub-picture on which a recommended word needs to be displayed.
S204: and displaying the M recommended words in the M target sub-pictures according to the mapping relation between the preset heat value of the recommended words and the significant value of the target sub-pictures.
Specifically, the mapping relationship between the heat value of the recommended word and the saliency value of the target sub-picture may be that the highest heat value corresponds to the highest saliency value, the next highest heat value corresponds to the next highest saliency value, and so on. Based on the mapping relationship, displaying M recommended words in M target sub-pictures specifically includes: and displaying the M recommended words in the M target sub-pictures, wherein the larger the significance value of the target sub-picture is, the larger the heat value of the recommended words displayed in the target sub-picture is. That is, the recommended word with the highest heat value is displayed in the target sub-picture with the highest significance value, the recommended word with the second highest heat value is displayed in the target sub-picture with the second highest significance value, and so on until all the recommended words are displayed. By the implementation mode, the recommended word with the highest heat value can be displayed in the target sub-picture most concerned by the user, the user can determine the interactive information needing to be input more conveniently, and the user experience is further improved.
It should be noted that there may be other ways for the mapping relationship between the heat value of the recommended word and the significant value of the target sub-picture, which is not limited in this embodiment of the present invention. For example, the highest significant value may correspond to a plurality of heat values, and a plurality of recommended words may be displayed in the target sub-picture with the highest significant value.
The terminal equipment can display the recommended words at any position of the target sub-picture. Alternatively, the target sub-picture may be rectangular in shape. The recommended word can be displayed along the length direction of the target sub-picture, namely, the recommended word is displayed transversely; the recommended word can also be displayed along the width direction of the target sub-picture, namely, the recommended word is displayed vertically; the recommended word may also be displayed along a diagonal direction of the target sub-picture, i.e., the recommended word is displayed diagonally.
The word sizes and colors of the characters of the M recommended words displayed in the M target sub-pictures can be completely the same or different. For example, the word size of the recommended word displayed in the target sub-picture with the largest saliency value may be larger than the word sizes of the recommended words displayed in other target sub-pictures, and the color may be different from that of the other recommended words. Further, the word sizes and colors of the characters of the M recommended words displayed in the M target sub-pictures may be different from each other.
Each recommended word may be displayed with a preset animation effect, for example, an animation effect in which the recommended word fluctuates in a wave shape.
The following describes the processes of S202 to S204 with a specific example. Suppose that 5 recommended words are provided, and the recommended words are respectively: tv shows, what news is there today, XXX videos, what is the weather today, and volume plus +. The heat value is sorted from big to small as follows: "tv drama" heat value, "what news is there today," XXX video, "what is today," what is the weather "heat value, and" volume plus + "heat value. FIG. 3A is a diagram illustrating a background picture of the current interactive interface in the embodiment shown in FIG. 2. Fig. 3B is a schematic diagram of the saliency map and the determined sub saliency map of the background picture shown in fig. 3A. Fig. 3C is a schematic diagram illustrating M recommended words displayed in M target sub-pictures. Using the Itti algorithmThe saliency map of fig. 3A is determined. Fig. 3B shows the saliency map of fig. 3A, for example, the background picture is divided into 3 × 3 sub-pictures on average, which correspond to 9 sub-saliency maps, in fig. 3B, the sub-saliency maps are respectively from bottom to top and from left to right: i is0、I1、I2、I3、I4、I5、I6、I7And I8. The size of the background picture is 333 × 268 pixels, and the size of each sub-picture is 111 × 89 pixels. And determining the sum of the significant values of all pixel points in each sub-significant map as the significant value of the sub-picture corresponding to the sub-significant map. In FIG. 3B, sub-saliency map I0-I8The significant values of the corresponding sub-pictures are respectively
And taking 5 sub-pictures with larger significance values as target sub-pictures. Then the 5 target sub-pictures are respectively: sub-saliency map I4Corresponding sub-picture, sub-saliency map I5Corresponding sub-picture, sub-saliency map I2Corresponding sub-picture, sub-saliency map I1Corresponding sub-picture and sub-saliency map I7The corresponding sub-picture. As shown in fig. 3C, the recommended word "drama" with the highest popularity value is displayed in the sub-saliency map I4On the corresponding sub-picture, the recommended word "what news is there today" with the second highest heat value is displayed on the sub-saliency map I5And (4) on the corresponding sub-picture, analogizing in sequence, and displaying all recommended words.
FIG. 4 is a diagram illustrating a variation of the user interface in the embodiment shown in FIG. 2. As shown in fig. 4, assume that the user is watching a television program through the television 41, and at this time, the interface of the television is shown as interface 43. The user triggers an interactive command via the remote control 42. After detecting the interactive instruction, the television executes S201-S204, and finally displays the interactive interface 44. In the interactive interface 44, the background picture is displayed in a partial area of the television screen — the background picture and the recommended word are displayed below the "long-press speech-like" display area. The user can input speech information via the microphone of the remote control or the microphone of the television set, at which time the interface 45 is displayed on the television set. And after receiving the voice information input by the user, the television sends the voice information to the server. And after the server performs voice recognition and semantic analysis, sending text information corresponding to the voice information input by the user to the television. The television displays text information corresponding to the voice information input by the user on the interface 46. For example, in interface 45, the user enters the speech "tv series", and in interface 46, the tv will display the text of "tv series". At the same time, the server sends the search result determined according to the voice information input by the user to the tv, and the tv can also simultaneously display the information returned by the server on the interface 46, for example, tv series 1-4 searched by the server. Next, the user can play the corresponding tv series by clicking the "play" button. In the process, because the television displays the recommended words determined based on the historical interaction data of the user at different positions of the background picture in the interface 44, the user can more quickly determine the interaction information which the user wants to input, and the interaction efficiency is improved.
The method for displaying recommended words provided by the embodiment of the invention comprises the steps of obtaining M recommended words and the heat value of each recommended word when an interactive instruction triggered by a user is detected, wherein the M recommended words are determined according to historical interactive data of the user, dividing a background picture of a current interactive interface into N sub-pictures, determining the significant value of each sub-picture, wherein the significant value of each sub-picture is used for representing the attention degree of the sub-picture by the user, selecting M sub-pictures with larger significant values from the N sub-pictures as target sub-pictures according to the significant values of the N sub-pictures, displaying the M recommended words in the M target sub-pictures according to the mapping relation between the preset heat value of the recommended word and the significant value of the target sub-pictures, on one hand, determining the recommended words and the heat value of each recommended word according to the historical interactive data of the user is realized, on the other hand, after the background picture is divided into a plurality of sub-pictures, M sub-pictures with high user attention degrees are selected as target sub-pictures, and finally M recommended words are displayed in the M target sub-pictures based on the mapping relation between the heat value and the significance value of the target sub-pictures. The recommended word display method provided by the embodiment of the invention determines different recommended words for different users, can display each recommended word in different areas of the background picture based on the characteristics of the background picture, has higher accuracy of the recommended word due to the fact that the display content is determined based on the personalized interactive data of the users, and is more flexible due to the fact that the display position is determined based on the characteristics of the background picture and the heat of the recommended word, thereby improving the user experience.
Fig. 5 is a schematic structural diagram of a recommended word display device according to an embodiment of the present invention. As shown in fig. 5, the recommended word display apparatus provided in the embodiment of the present invention includes the following modules: an acquisition module 51, a first determination module 52, a selection module 53 and a display module 54.
The obtaining module 51 is configured to obtain M recommended words and a heat value of each recommended word when an interaction instruction triggered by a user is detected.
The M recommended words are determined according to historical interaction data of the user, and M is an integer greater than or equal to 1.
Specifically, the obtaining module 51 obtains the recommended word and the heat value of the recommended word in the following two implementation manners:
in a first implementation manner, the obtaining module 51 is specifically configured to: and acquiring the M recommended words and the heat value of each recommended word from the server.
In a second implementation manner, the obtaining module 51 is specifically configured to: determining a candidate recommended word set of the user according to historical interaction data of the user, and determining a heat value of the candidate recommended word; and selecting M candidate recommended words with larger heat values from the candidate recommended word set as the M recommended words according to the heat values of the candidate recommended words. In this implementation manner, the obtaining module 51 determines the heat value of each candidate recommended word in the candidate recommended word set, including: according to the formula
And determining the heat value of each candidate recommended word. Wherein x is1Representing the number of occurrences, x, of the candidate referral in the first period2Representing the number of occurrences of the candidate recommended word in a second period of time that is closer than the first period of timeAnd the current time is the time for acquiring the interactive instruction triggered by the user.
The first determining module 52 is configured to divide the background picture of the current interactive interface into N sub-pictures, and determine a saliency value of each sub-picture.
Wherein the significance value of the sub-picture is used for representing the degree of attention of the sub-picture by a user, and N is an integer greater than or equal to M.
Optionally, the recommended word display device provided in the embodiment of the present invention further includes: and the second determination module is used for determining the saliency map of the background picture and the sum of the saliency values of all the pixel points in the saliency map according to the Itti algorithm. Then, the first determining module 52 is specifically configured to: averagely dividing the background picture into P × P sub-pictures, and determining a sub-saliency map of each sub-picture according to the saliency map of the background picture; and determining the quotient of the sum of the significant values of all the pixel points in each sub-significant image and the sum of the significant values of all the pixel points in the significant image of the background image as the significant value of each sub-image. Wherein P represents the number of sub-pictures in the length direction and width direction of the saliency map, (P-1)2＜M≤P2And P is an integer, P ═ N.
And a selecting module 53, configured to select M sub-pictures with larger significance values from the N sub-pictures as target sub-pictures according to the magnitudes of the significance values of the N sub-pictures.
And the display module 54 is configured to display the M recommended words in the M target sub-pictures according to a mapping relationship between the preset heat value of the recommended word and the significant value of the target sub-picture.
Optionally, the display module 54 is specifically configured to: and displaying the M recommended words in the M target sub-pictures, wherein the larger the significance value of the target sub-picture is, the larger the heat value of the recommended words displayed in the target sub-picture is.
The recommended word display device provided by the embodiment of the invention is provided with an acquisition module for acquiring M recommended words and a heat value of each recommended word when an interaction instruction triggered by a user is detected, wherein the M recommended words are determined according to historical interaction data of the user, a first determination module for dividing a background picture of a current interaction interface into N sub-pictures and determining a significant value of each sub-picture, wherein the significant value of the sub-pictures is used for indicating the degree of attention of the sub-pictures by the user, a selection module for selecting M sub-pictures with larger significant values from the N sub-pictures as target sub-pictures according to the significant values of the N sub-pictures, a display module for displaying the M recommended words in the M target sub-pictures according to a mapping relation between the preset heat value of the recommended word and the significant value of the target sub-pictures, and on one hand, the determination of the recommended words and the heat value of each recommended word according to the historical interaction data of the user is realized, on the other hand, after the background picture is divided into a plurality of sub-pictures, M sub-pictures with high user attention degrees are selected as target sub-pictures, and finally M recommended words are displayed in the M target sub-pictures based on the mapping relation between the heat value and the significance value of the target sub-pictures. The recommended word display device provided by the embodiment of the invention determines different recommended words for different users, can display each recommended word in different areas of a background picture based on the characteristics of the background picture, has higher accuracy of the recommended word due to the fact that the display content is determined based on the personalized interactive data of the users, and is more flexible due to the fact that the display position is determined based on the characteristics of the background picture and the heat of the recommended word, thereby improving the user experience.
Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A recommended word display method, comprising:
when an interactive instruction triggered by a user is detected, acquiring M recommended words and the heat value of each recommended word; the M recommended words are determined according to historical interaction data of the user, and M is an integer greater than or equal to 1;
dividing a background picture of a current interactive interface into N sub-pictures, and determining a significant value of each sub-picture; wherein the significance value of the sub-picture is used for representing the degree of attention of the sub-picture by a user, and N is an integer greater than or equal to M;
according to the sizes of the significant values of the N sub-pictures, M sub-pictures with larger significant values are selected from the N sub-pictures as target sub-pictures;
displaying the M recommended words in the M target sub-pictures according to a mapping relation between a preset heat value of the recommended words and a significant value of the target sub-pictures, displaying different recommended words at different target sub-picture positions in the background picture, wherein the larger the significant value of the target sub-pictures is, the larger the heat value of the recommended words displayed in the target sub-pictures is.
2. The method according to claim 1, wherein the displaying the M recommended words in the M target sub-pictures specifically includes:
displaying the M recommended words in the M target sub-pictures, wherein the larger the significance value of the target sub-picture is, the larger the heat value of the recommended words displayed in the target sub-picture is.
3. The method according to claim 1 or 2, wherein the obtaining of the M recommended words and the heat value of each recommended word comprises:
and acquiring the M recommended words and the heat value of each recommended word from a server.
4. The method according to claim 1 or 2, wherein the obtaining of the M recommended words and the heat value of each recommended word comprises:
determining a candidate recommended word set of the user according to the historical interaction data of the user, and determining the heat value of the candidate recommended word;
and selecting M candidate recommended words with larger heat values from the candidate recommended word set as the M recommended words according to the heat values of the candidate recommended words.
5. The method of claim 4, wherein the determining the popularity value of the candidate recommended word comprises:
according to the formula
Determining the heat value of each candidate recommended word; wherein x is1Representing the number of occurrences, x, of the candidate recommended word in a first period2And representing the frequency of occurrence of the candidate recommended word in a second time period, wherein the second time period is closer to the current time than the first time period, and the current time is the time for acquiring the interactive instruction triggered by the user.
6. A recommended word display device, comprising:
the acquisition module is used for acquiring M recommended words and the heat value of each recommended word when an interactive instruction triggered by a user is detected; the M recommended words are determined according to historical interaction data of the user, and M is an integer greater than or equal to 1;
the first determining module is used for dividing the background picture of the current interactive interface into N sub-pictures and determining the significant value of each sub-picture; wherein the significance value of the sub-picture is used for representing the degree of attention of the sub-picture by a user, and N is an integer greater than or equal to M;
a selection module, configured to select M sub-pictures with larger significance values from the N sub-pictures as target sub-pictures according to the significance values of the N sub-pictures;
the display module is configured to display the M recommended words in the M target sub-pictures according to a mapping relationship between a preset heat value of the recommended word and a significant value of a target sub-picture, display different recommended words at different target sub-picture positions in the background picture, and increase the heat value of the recommended word displayed in the target sub-picture when the significant value of the target sub-picture is larger.
7. The apparatus of claim 6, wherein the display module is specifically configured to:
displaying the M recommended words in the M target sub-pictures, wherein the larger the significance value of the target sub-picture is, the larger the heat value of the recommended words displayed in the target sub-picture is.
8. The apparatus according to claim 6 or 7, wherein the obtaining module is specifically configured to:
and acquiring the M recommended words and the heat value of each recommended word from a server.
9. The apparatus according to claim 6 or 7, wherein the obtaining module is specifically configured to:
determining a candidate recommended word set of the user according to the historical interaction data of the user, and determining the heat value of the candidate recommended word;
and selecting M candidate recommended words with larger heat values from the candidate recommended word set as the M recommended words according to the heat values of the candidate recommended words.
10. The apparatus of claim 9, wherein the obtaining module determines the popularity value of the candidate recommended word, and comprises:
according to the formula
Determining the heat value of each candidate recommended word; wherein x is1Representing the number of occurrences, x, of the candidate recommended word in a first period2And representing the frequency of occurrence of the candidate recommended word in a second time period, wherein the second time period is closer to the current time than the first time period, and the current time is the time for acquiring the interactive instruction triggered by the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710587081.3A CN107391661B (en) | 2017-07-18 | 2017-07-18 | Recommended word display method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710587081.3A CN107391661B (en) | 2017-07-18 | 2017-07-18 | Recommended word display method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107391661A CN107391661A (en) | 2017-11-24 |
| CN107391661B true CN107391661B (en) | 2020-06-16 |
Family
ID=60339959
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710587081.3A Active CN107391661B (en) | 2017-07-18 | 2017-07-18 | Recommended word display method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107391661B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108921855A (en) * | 2018-05-31 | 2018-11-30 | 上海爱优威软件开发有限公司 | Image processing method and system based on information |
| CN110750682B (en) * | 2018-07-06 | 2022-08-16 | 武汉斗鱼网络科技有限公司 | Title hot word automatic metering method, storage medium, electronic equipment and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103198072A (en) * | 2012-01-06 | 2013-07-10 | 腾讯科技（深圳）有限公司 | Method and device for mining and recommendation of popular search word |
| CN103577993A (en) * | 2012-08-07 | 2014-02-12 | 阿里巴巴集团控股有限公司 | Color selecting method and device |
| CN104536964A (en) * | 2014-11-17 | 2015-04-22 | 北京国双科技有限公司 | A network data displaying method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101537174B1 (en) * | 2013-12-17 | 2015-07-15 | 가톨릭대학교 산학협력단 | Method for extracting salient object from stereoscopic video |
-
2017
- 2017-07-18 CN CN201710587081.3A patent/CN107391661B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103198072A (en) * | 2012-01-06 | 2013-07-10 | 腾讯科技（深圳）有限公司 | Method and device for mining and recommendation of popular search word |
| CN103577993A (en) * | 2012-08-07 | 2014-02-12 | 阿里巴巴集团控股有限公司 | Color selecting method and device |
| CN104536964A (en) * | 2014-11-17 | 2015-04-22 | 北京国双科技有限公司 | A network data displaying method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107391661A (en) | 2017-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109919087B (en) | Video classification method, model training method and model training device | |
| KR20210015951A (en) | Image processing method and apparatus, electronic device, and storage medium | |
| EP3796189A1 (en) | Video retrieval method, and method and apparatus for generating video retrieval mapping relationship | |
| CN111445902B (en) | Data collection method, device, storage medium and electronic equipment | |
| CN110070063B (en) | Target object motion recognition method and device and electronic equipment | |
| US10701301B2 (en) | Video playing method and device | |
| US20200404378A1 (en) | Video-based information acquisition method and device | |
| CN111405339B (en) | Split screen display method, electronic equipment and storage medium | |
| JP2019529992A (en) | Display device and control method thereof | |
| US20230316529A1 (en) | Image processing method and apparatus, device and storage medium | |
| CN110691281A (en) | Video playing processing method, terminal device, server and storage medium | |
| CN111078011A (en) | Gesture control method and device, computer readable storage medium and electronic equipment | |
| CN110177295A (en) | Processing method, device and the electronic equipment that subtitle crosses the border | |
| CN113076048A (en) | Video display method and device, electronic equipment and storage medium | |
| CN112380929A (en) | Highlight segment obtaining method and device, electronic equipment and storage medium | |
| CN107391661B (en) | Recommended word display method and device | |
| CN111768377A (en) | Image color evaluation method and device, electronic equipment and storage medium | |
| US20170161871A1 (en) | Method and electronic device for previewing picture on intelligent terminal | |
| CN110868632A (en) | Video processing method and device, storage medium and electronic equipment | |
| CN112785669B (en) | Virtual image synthesis method, device, equipment and storage medium | |
| CN112866577B (en) | Image processing method and device, computer readable medium and electronic equipment | |
| CN111274449B (en) | Video playing method, device, electronic equipment and storage medium | |
| CN112256976A (en) | Matching method and related device | |
| CN112312208A (en) | Multimedia information processing method and device, storage medium and electronic equipment | |
| CN112182299B (en) | Method, device, equipment and medium for acquiring highlight in video |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |