CN117131241B - Search object recommendation method, electronic device and computer readable storage medium - Google Patents

Search object recommendation method, electronic device and computer readable storage medium Download PDF

Info

Publication number
CN117131241B
CN117131241B CN202310143444.XA CN202310143444A CN117131241B CN 117131241 B CN117131241 B CN 117131241B CN 202310143444 A CN202310143444 A CN 202310143444A CN 117131241 B CN117131241 B CN 117131241B
Authority
CN
China
Prior art keywords
search
word
similarity
candidate
words
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
Application number
CN202310143444.XA
Other languages
Chinese (zh)
Other versions
CN117131241A (en
Inventor
肖芸榕
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Honor Device Co Ltd
Original Assignee
Honor Device Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Honor Device Co Ltd filed Critical Honor Device Co Ltd
Priority to CN202310143444.XA priority Critical patent/CN117131241B/en
Publication of CN117131241A publication Critical patent/CN117131241A/en
Application granted granted Critical
Publication of CN117131241B publication Critical patent/CN117131241B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06F16/903—Querying
    • G06F16/9035—Filtering based on additional data, e.g. user or group profiles
    • 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
    • G06F16/903—Querying
    • G06F16/9038—Presentation of query results
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
    • G06F3/0487—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Human Computer Interaction (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The application provides a search object recommending method, electronic equipment and a computer readable storage medium. The search object recommending method comprises the following steps: acquiring search words; determining first similarity between the search word and the candidate word according to the bipartite graph, and determining association words of the search word from the candidate word according to the first similarity; and recommending target search objects according to the association words. The two-part graph comprises a first set, a second set and a plurality of edges, wherein the first set comprises candidate words, the second set comprises search objects, each edge is used for connecting the mutually-associated candidate words and the search objects, the candidate words of the same search behavior are mutually associated with the search objects corresponding to the click behavior, and the candidate words of the same search behavior represent the candidate words detected from the detection of the input behavior to the detection of the click behavior. According to the bipartite graph, the association words of more characters similar to the search words of fewer characters can be determined, and effective search objects are recommended for the user under the condition that the user inputs fewer characters, so that the search efficiency is improved.

Description

Search object recommendation method, electronic device and computer readable storage medium
Technical Field
The present application relates to the field of information processing, and in particular, to a search object recommendation method, an electronic device, and a computer readable storage medium.
Background
When the electronic equipment detects that the user inputs the search word, the electronic equipment recommends a search object for the user according to the search word, so that the user input is reduced. In the prior art, an electronic device generally determines a search object according to the semantics and grammar of a search word, and a user needs to input more characters to determine the search intention of the user. The search intention of the user cannot be determined when the user inputs fewer characters, more irrelevant search objects are recommended for the user or search objects cannot be recommended for the user, namely effective search objects cannot be recommended for the user under the condition that the user inputs fewer characters, so that the search efficiency is reduced.
Disclosure of Invention
The application provides a search object recommending method, electronic equipment and a computer readable storage medium, which solve the problem that effective search objects cannot be recommended for users when the users input fewer characters in the prior art.
In order to achieve the above purpose, the application adopts the following technical scheme:
In a first aspect, a search object recommendation method is provided, including: acquiring search words; determining a first similarity between the search word and a candidate word according to a bipartite graph, wherein the bipartite graph is generated according to historical search behaviors, the bipartite graph comprises a first set, a second set and edges, the first set comprises the candidate word, the second set comprises a search object, the values of the edges represent the association degree between the candidate word and the search object which are mutually associated, the candidate words of the same search behavior in the historical search behaviors are mutually associated with the search object corresponding to the clicking behavior, and the candidate words of the same search behavior represent all the candidate words detected between the detection of an input behavior and the detection of the clicking behavior; determining association words of the search words from the candidate words according to the first similarity; and recommending a target search object according to the association word.
In the above embodiment, the candidate words of the same search behavior in the two graphs include candidate words with few characters to many characters, and because the candidate words of the same search behavior in the two graphs are all associated with the search object corresponding to the click behavior, the first similarity between the search word and the candidate word is determined according to the two graphs, the associated word with more characters similar to the search word with fewer characters can be quickly determined, and then the search intention of the user can be identified under the condition that the user inputs fewer characters according to the associated word recommendation target object, so that the effective search object is recommended for the user, and the search efficiency is further improved.
In an embodiment, the value of the edge is determined according to the number of clicks of the candidate word, where the number of clicks of the candidate word indicates the number of clicks of the corresponding search object according to the candidate word. The search object corresponding to the candidate word click refers to that the user directly clicks the search object after inputting the candidate word. According to the click times of the candidate words, the value of the edge is determined, so that the correlation between the candidate words and the search object can be improved, and the correlation between the value of the edge in the bipartite graph and the search behavior of the user is improved.
In an embodiment, the determining the first similarity between the search term and the candidate term according to the bipartite graph includes:
And determining the first similarity between the search word and the candidate word according to the values of all edges corresponding to the search word, the values of all edges corresponding to the candidate word and the second similarity, wherein the second similarity is the similarity between the search object associated with the search word and the search object associated with the candidate word. And when the similarity between the search word and the candidate word is calculated, the second similarity of the search object associated with the search word and the candidate word is considered, so that the first similarity between the search word and the candidate word can be better evaluated, and the correlation between the first similarity and the search word and the correlation between the first similarity and the candidate word are improved.
In an embodiment, the determining the second similarity according to the values of all edges corresponding to the search term, the values of all edges corresponding to the candidate term, and the second similarity includes: and carrying out iterative operation according to the values of all sides corresponding to the search word, the values of all sides corresponding to the candidate word, the second similarity and a preset iterative formula to obtain the first similarity between the search word and the candidate word. By combining all sides of the candidate words and all sides of the search words and performing iterative operation, the first similarity is obtained, the indirect relation search words and the candidate words can be connected, and the relevance of the obtained association words and the search words is improved.
In an embodiment, before the performing an iterative operation according to the values of all edges corresponding to the search term, the values of all edges corresponding to the candidate term, the second similarity, and a preset iterative formula, the method further includes: according to the search object associated with the search word and the search object associated with the candidate word, determining a first initial similarity between the search word and the candidate word, taking the first initial similarity as an initial value of the iterative operation, or according to the candidate word corresponding to the search object, determining a second initial similarity between the search object associated with the search word and the search object associated with the candidate word, and taking the second initial similarity as an initial value of the iterative operation. The first initial similarity is determined through the search object associated with the search word and the search object associated with the candidate word, so that the correlation between the obtained first initial similarity and the search word can be improved, and the second initial similarity is determined through the candidate word associated with the search object, so that the correlation between the obtained second initial similarity and the search object can be improved.
In an embodiment, the determining the associated word of the search word from the candidate words according to the first similarity includes: and using the candidate words with the first similarity larger than a preset value as the association words of the search words, so that the association words with higher association degrees with the search words can be obtained.
In an embodiment, the number of the associated words is a plurality, and the recommending the target search object according to the associated words includes: determining the search object associated with each associated word; and taking the search object with the relevance larger than a preset value as the target search object and outputting the target search object, so that an effective target search object can be recommended for a user, and the search path is shortened.
In one embodiment, after recommending the target search object according to the associative word, the method further includes: and updating the bipartite graph according to the target search object clicked by the user, so that the data for constructing the bipartite graph can be enriched, and the bipartite graph is optimized.
In a second aspect, there is provided a search object recommending apparatus including:
the acquisition module is used for acquiring the search word;
The computing module is used for determining first similarity between the search word and the candidate word according to a bipartite graph, the bipartite graph is generated according to historical search behaviors, the bipartite graph comprises a first set, a second set and a plurality of edges, the first set comprises the candidate word, the second set comprises a search object, each edge is used for connecting the candidate word and the search object which are mutually related, and the value of each edge represents the degree of association between the candidate word and the search object which are mutually related; the candidate words of the same search behavior in the historical search behaviors are mutually associated with the search objects corresponding to the click behaviors, and the candidate words of the same search behavior represent all candidate words detected from the detection of the input behavior to the detection of the click behavior;
A determining module, configured to determine, from the candidate words, an associative word of the search word according to the first similarity;
And the recommending module is used for recommending the target search object according to the association word.
In an embodiment, the value of the edge is determined according to the number of clicks of the candidate word, where the number of clicks of the candidate word indicates the number of clicks of the corresponding search object according to the candidate word.
In one embodiment, the computing module is specifically configured to:
And determining the first similarity between the search word and the candidate word according to the values of all edges corresponding to the search word, the values of all edges corresponding to the candidate word and the second similarity, wherein the second similarity is the similarity between the search object associated with the search word and the search object associated with the candidate word.
In an embodiment, the second similarity is determined by the first similarity, and the calculation module is specifically configured to:
and carrying out iterative operation according to the values of all sides corresponding to the search word, the values of all sides corresponding to the candidate word, the second similarity and a preset iterative formula to obtain the first similarity between the search word and the candidate word.
In an embodiment, the computing module is further to:
According to the search object associated with the search word and the search object associated with the candidate word, determining a first initial similarity between the search word and the candidate word, taking the first initial similarity as an initial value of the iterative operation, or according to the candidate word corresponding to the search object, determining a second initial similarity between the search object associated with the search word and the search object associated with the candidate word, and taking the second initial similarity as an initial value of the iterative operation.
In an embodiment, the determining module is specifically configured to:
and taking the candidate words with the first similarity larger than a preset value as the association words of the search words.
In an embodiment, the number of the association words is a plurality, and the recommendation module is specifically configured to:
Determining the search object associated with each associated word;
and taking the search object with the relevance being larger than a preset value as the target search object and outputting the target search object.
In an embodiment, the computing module is further to:
And updating the bipartite graph according to the target search object clicked by the user.
In a third aspect, an electronic device is provided, comprising a processor configured to execute a computer program stored in a memory, to implement the search object recommendation method as described in the first aspect above.
In a fourth aspect, there is provided a computer readable storage medium storing a computer program which, when executed by a processor, implements the search object recommendation method according to the first aspect described above.
In a fifth aspect, a chip is provided, the chip comprising a processor, the processor being coupled to a memory, the processor executing a computer program or instructions stored in the memory to implement the search object recommendation method according to the first aspect described above.
In a sixth aspect, there is provided a computer program product for, when run on an electronic device, causing the electronic device to perform the search object recommendation method according to the first aspect described above.
It will be appreciated that the advantages of the second to sixth aspects may be found in the relevant description of the first aspect, and are not described here again.
Drawings
Fig. 1 is a flow chart of a search object recommending method according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a scenario in which a user inputs a search term according to an embodiment of the present application;
FIG. 3 is a schematic diagram of another scenario in which a user inputs a search term according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a bipartite graph according to an embodiment of the present application;
FIG. 5 is an application scenario diagram of a method for recommending search objects according to an embodiment of the present application;
FIG. 6 is another application scenario diagram of a method for recommending search objects according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application;
Fig. 8 is a software architecture diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth such as the particular system architecture, techniques, etc., in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.
It should be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
As used in the present description and the appended claims, the term "if" may be interpreted as "when..once" or "in response to a determination" or "in response to detection" depending on the context. Similarly, the phrase "if a determination" or "if a [ described condition or event ] is detected" may be interpreted in the context of meaning "upon determination" or "in response to determination" or "upon detection of a [ described condition or event ]" or "in response to detection of a [ described condition or event ]".
In addition, in the description of the present application, the terms "first," "second," and the like are used merely to distinguish between descriptions and are not to be construed as indicating or implying relative importance.
Reference in the specification to "one embodiment" or "some embodiments" or the like means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the application. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," and the like in the specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments" unless expressly specified otherwise.
In the prior art, the electronic device needs more characters input by the user to determine the search intention of the user, and recommends a search object for the user, so that the search efficiency is lower.
To this end, the present application provides a search object recommendation method including: and obtaining the search word, and determining the first similarity between the search word and the candidate word according to the bipartite graph. The method comprises the steps that a two-part graph is generated according to historical search behaviors, the two-part graph comprises a first set, a second set and edges, the first set comprises candidate words, the second set comprises search objects, the values of the edges represent the association degrees between the mutually associated candidate words and the search objects, the candidate words of the same search behavior in the historical search behaviors are mutually associated with the search objects corresponding to clicking behaviors, and the candidate words of the same search behavior represent all the candidate words detected from the detection of input behaviors to the detection of clicking behaviors; determining association words of the search word from the candidate words according to the first similarity; and recommending target search objects according to the association words.
Since the candidate words of the same search behavior represent all the candidate words detected between the detection of the input behavior and the detection of the click behavior, the same search behavior includes the candidate words corresponding to the click behavior and the candidate words of fewer characters input by the user before the detection of the click behavior. Because the candidate words of the same search behavior in the two graphs are mutually associated with the search objects corresponding to the clicking behavior, the first similarity between the search word and the candidate word is determined according to the two graphs, the associated word with more characters similar to the search word with fewer characters can be rapidly determined, then the target object is recommended according to the associated word, the search intention of the user can be identified under the condition that the user inputs fewer characters, the effective search object is recommended for the user, and the search efficiency is further improved.
The search object recommendation method provided by the embodiment of the application is exemplified below.
The search object recommendation method provided by the embodiment of the application can be applied to electronic equipment and also can be applied to a server in communication with the electronic equipment. The search object recommending method provided by the embodiment of the application can be applied to electronic equipment for example, and is described.
The electronic device in the embodiments of the present application may be a mobile phone, a tablet computer, a handheld computer, a Personal Digital Assistant (PDA), an Augmented Reality (AR) device, a media player, a wearable device, or a device that can be held/operated by one hand, and the specific form/type of the electronic device is not particularly limited. The electronic device includes but is not limited to a mounted deviceHong Mongolian System (Harmony OS) or other operating system devices.
As shown in fig. 1, the search object recommendation method provided in an embodiment of the present application includes:
s101: search terms are obtained.
Specifically, the characters input by the user on the search interface, which are detected by the electronic device, are search words.
The search interface may be an interface for searching for content such as applications, services, games, web pages, pictures, videos, and the like. The search term may be one or more characters of chinese, english, numbers, etc.
In one embodiment, the electronic device uses the character currently input by the user as a search term when detecting a search instruction of the user. For example, as shown in FIG. 2, at the search interface of an application, the user clicks the "search" control after the search interface inputs "motion". The electronic device uses the motion as a search term when detecting that a user clicks a search control.
In another embodiment, the electronic device uses the character currently displayed on the search interface as a search term. For example, as shown in fig. 3, in the search interface of the application program, the user sequentially inputs the real, and the electronic device takes "h" as a search term after detecting the user input "h", takes "he" as a search term after detecting the user input "e", takes "hea" as a search term after detecting the user input "a", and takes "real" as a search term after detecting the user input "l".
S102: determining a first similarity between the search word and a candidate word according to a bipartite graph, wherein the bipartite graph is generated according to historical search behaviors, the bipartite graph comprises a first set, a second set and a plurality of edges, the first set comprises the candidate word, the second set comprises a search object, each edge is used for connecting the candidate word and the search object which are mutually related, and the value of each edge represents the degree of association between the candidate word and the search object which are mutually related; the candidate words of the same search behavior in the historical search behaviors are mutually associated with the search objects corresponding to the clicking behaviors, and the candidate words of the same search behavior represent all the candidate words detected from the detection of the input behavior to the detection of the clicking behavior.
Specifically, the bipartite graph also refers to a bipartite graph, which is generated in advance according to a historical search behavior of a preset period (for example, one week, one month, one year, etc.). The historical search behavior is the behavior that a user searches by inputting candidate words into a search interface and clicks a search object recommended by the electronic equipment. The candidate words in the first set are search words detected by the electronic device in the historical search behavior, and the search objects in the second set are objects clicked by the user detected by the electronic device in the historical search behavior. In the history searching behavior, a candidate word is input by a user, the electronic equipment recommends a searching object for the user, the user clicks one of the recommended searching objects, the candidate word is correlated with the searching object clicked by the user, and a line segment connecting the correlated candidate word and the searching object is an edge. The historical search behavior can be the historical search behavior on the current electronic equipment or the historical search behavior reported by a plurality of electronic equipment collected by a server. The two-part map may be generated by the electronic device or may be generated by the server and then acquired from the server by the electronic device.
Candidate words for the same search behavior represent all candidate words detected between the detection of an input behavior and the detection of a click behavior. Specifically, when the electronic device detects that the user opens the search interface and does not input the search interface, if the input behavior of the user is detected, determining that the search behavior starts, recording a character currently displayed on the search interface, taking the recorded character as a candidate word of the current search behavior, and when the click behavior of the user is detected, ending the search behavior, wherein the candidate word recorded from the detection of the input behavior to the detection of the click behavior is a candidate word of the same search behavior.
One search behavior corresponds to one click behavior, so that candidate words of the same search behavior correspond to one click behavior, and a search object corresponding to the click behavior refers to an object clicked by the click behavior, so that candidate words of the same search behavior correspond to one search object.
For example, when the electronic device detects that the user opens the search interface, the electronic device starts to monitor the input behavior of the user, if the user input of "fortune" is detected, determines that the search behavior starts, uses "fortune" as a candidate word and records, if the user continues to input "movement" is detected, uses "movement" as a candidate word and records, if the user continues to input "health" is detected, uses "movement health" as a candidate word and records, and if the user continues to input "health" is detected, uses "movement health" as a candidate word and records. Then, if the operation of clicking the 'sports health' by the user is detected, the end of the search behavior is determined, the 'sports', 'sports health' are used as candidate words of the same search behavior, the candidate words of the same search behavior correspond to one click behavior, and the search object corresponding to the click behavior is 'sports health'.
If the user inputs 'sports health', the input information is deleted, namely, after the input behavior is detected, the clicking behavior is not detected, and no character exists on the current search interface, the electronic equipment determines that no candidate word corresponding to the search behavior exists, and starts to monitor the next search behavior. For example, in the case where an input behavior of the user is detected, it is determined that the next search behavior starts, and the corresponding candidate word is recorded.
For another example, when the electronic device detects that the user opens the search interface, it starts to monitor the input behavior of the user, if the user inputs "h", it takes "h" as a candidate word and records, if the user continues to input "e", it takes "he" as a candidate word and records, if the user continues to input "e", it takes "hee" as a candidate word and records, if the user deletes "e" and then inputs "a", it takes "hea" as a candidate word and records, and if the user continues to input "l" is detected, it takes "real" as a candidate word and records. Then, if the operation of clicking "HEALTH APP" by the user is detected, the "h", "he", "hee", "hea" and "seal" are used as candidate words of the same search behavior, the candidate words of the same search behavior correspond to one click behavior, and the search object corresponding to the click behavior is "HEALTH APP".
For another example, when the electronic device detects that the user opens the search interface, it starts to monitor the input behavior of the user, if the operation of pasting a character by the user is detected, the pasting operation is performed, the pasted character "hea" is displayed on the display interface, the "hea" is used as a candidate word and recorded, and if the user continues to input "l" is detected, the "heal" is used as a candidate word and recorded. Then, if the operation of clicking "HEALTH APP" by the user is detected, using "hea" and "seal" as candidate words of the same search behavior, where the candidate words of the same search behavior correspond to one click behavior, and the search object corresponding to the click behavior is "HEALTH APP".
Thus, candidate words for the same search activity include all candidate words displayed on the display interface prior to the user clicking on the search object, including both incomplete words entered by the user and incorrect words entered by the user.
After determining the search object associated with each candidate word in the historical search behavior, a bipartite graph can be constructed. Specifically, the candidate words of the same search behavior correspond to the search object clicked by the click behavior, that is, the candidate words of the same search behavior are all correlated with the search object clicked by the click behavior, and each edge in the two graphs is used for connecting the correlated candidate words and the search object, that is, the candidate words of the same search behavior have edges pointing to the corresponding search object. For example, "sports" and "sports health" are candidates for the same search behavior, and the corresponding search object is "sports health", and "sports", "sports health" and "sports health" all have edges pointing to "sports health". For another example, "h", "he", "hee", "hea", "heal" are candidates for the same search behavior, and the corresponding search object is "HEALTH APP", and "h", "he", "hee", "hea", "heal" each have an edge pointing to "HEALTH APP".
It will be appreciated that the same candidate word may correspond to a plurality of search actions, one for each search object, and thus the same candidate word may be associated with a plurality of search objects, and correspondingly, the same candidate word may have edges pointing to the plurality of search objects. For example, the search objects corresponding to "h", "he", "hea", "real" are "HEALTH APP", and the search objects corresponding to "h", "ha", "hap" are "happy", and then the candidate word "h" has an edge pointing to "HEALTH APP" and an edge pointing to "happy".
Candidate words pointing to the same search object may be words composed of different characters (e.g., chinese, english, numbers), respectively, or each candidate word may be composed of one or more characters. For example, in the history search behavior, when the electronic device detects that the user opens the search interface, it starts to monitor the input behavior of the user, if the user input of "fortune" is detected, it takes "fortune" as a candidate word and records, if the user input of "movement" is detected, it takes "sports" as a candidate word and records, if the user input of "health" is detected, it takes "sports health" as a candidate word and records, and if the user input of "health" is detected, it takes "sports health" as a candidate word and records. Then, if the operation of clicking "HEALTH APP" by the user is detected, the "fortune", the motion "," the motion exercise "and" the motion health "are candidates of the same search behavior, and the corresponding search objects are" HEALTH APP "," fortune ", the motion", "the motion exercise" and "the motion health" all have edges pointing to "HEALTH APP". Similarly, in another historical search behavior, "h", "he", "hea", "heal" are candidates for the same search behavior, and the corresponding search objects are "HEALTH APP", "h", "he", "hea", "heal" each have an edge pointing to "HEALTH APP".
And connecting the candidate words and the search words which are associated with each other by edges to obtain the bipartite graph. Illustratively, as shown in FIG. 4, the bipartite graph includes a set of candidate words and a set of search objects, each edge being used to connect the candidate words and the search objects that are associated with each other. The candidate words "fortune", "motion", "h", "he", "hea" in the candidate word set each have an edge pointing to the search object "HEALTH APP", the "h", "ha", "hap" each have an edge pointing to the search object "happy", and the "intelligent", "intelligent" each have an edge pointing to the search object "intelligent care".
The candidate words and the search objects connected by each edge in the two-part graph are associated with each other, and the values of the edges represent the association degree between the candidate words and the search objects.
In one embodiment, the value of the edge is determined according to the number of clicks of the candidate word, which represents the number of clicks of the corresponding search object according to the candidate word.
Specifically, in the history search behavior, in the case where the electronic device displays the search interface, after detecting the candidate word, the electronic device detects that the user continues to input the character, or detects the click behavior of the user. If the electronic equipment detects the candidate word and then detects that the user continues to input characters, the candidate word is the candidate word without clicking action in the current searching action. If the electronic equipment detects the candidate word and then detects the clicking action of the user, the candidate word is the candidate word with the clicking action in the current searching action. Thus, candidate words for the same search behavior include candidate words for which there is a click behavior, as well as candidate words for which there is no click behavior. One search behavior includes a plurality of candidate words, one search behavior corresponds to one click behavior and one search object, and the search behavior corresponding to the search object is also referred to as a search behavior of clicking the corresponding search object. And counting the times of displaying the candidate words on a search interface in the historical search behavior of clicking the search object for the edges connecting the candidate words and the search object, namely the times of detecting the candidate words. For the historical search behavior of the candidate word, counting the times of clicking the corresponding search object according to the candidate word, wherein the times of clicking the corresponding search object according to the candidate word are the times of clicking the candidate word as the candidate word with the click behavior, and the times of clicking the candidate word are the times of clicking the candidate word after the candidate word is detected and the times of detecting the search object by the user are the times of clicking the corresponding search object according to the candidate word.
In one embodiment, the value of the edge is equal to the number of clicks of the candidate word.
For example, as shown in fig. 4, in the search behavior in which the click behavior is clicking on "HEALTH APP" for the side connecting "h" and "HEALTH APP", the number of times the candidate word "h" is detected is 1500. In the search behavior in which "HEALTH APP" is clicked, after "h" is detected, it is not detected that the user continues to input characters, and the number of times that "HEALTH APP" is detected (i.e., the number of times "HEALTH APP" is clicked according to "h") is 200, the value of the edge is 200.
In another embodiment, the value of the edge is equal to the product of the number of times the candidate word is detected and the conversion rate after the smoothing process, which may be Wilson smoothing, is the ratio of the number of clicks to the number of times the candidate word is detected. The conversion rate can be corrected by smoothing the conversion rate, so that the conversion rate after smoothing with higher confidence is obtained.
For example, for a side connecting a candidate word and a search object, the value of the side is 300 if the number of times the candidate word is detected is 1500 words and the conversion rate after wilson smoothing is 0.2.
The more clicks the candidate word is, the higher the conversion rate after the corresponding smoothing process is. The product of the number of times of detecting the candidate word and the conversion rate after the smoothing is used for representing the value of the edge, so that the correlation between the candidate word and the search object can be improved, and the accuracy of representing the correlation degree by the value of the edge can be improved.
In other embodiments, the value of the edge may also be determined based on the number of clicks of the candidate word and the total number of clicks of the search object, including the number of clicks of the search object based on different candidate words.
It will be appreciated that candidate words, search objects, connection of edges, values of edges in the bipartite graph may also be stored in text.
The number of effective clicks is used as the value of the edge, and the number of clicks according to the candidate word is used as the value of the edge, so that the correlation between the candidate word and the search object can be improved, and the correlation between the value of the edge in the bipartite graph and the search behavior of the user can be improved.
In an embodiment, the search word is identical to one of the candidate words, and the first similarity between the search word and the candidate word is determined according to the values of all edges corresponding to the search word, the values of all edges corresponding to the candidate word, and the second similarity, where the second similarity is the similarity between the search object associated with the search word and the search object associated with the candidate word. Specifically, all edges corresponding to the search word are all edges connected with the search word, and all edges corresponding to the candidate word are all edges connected with the candidate word. The search object associated with the search word refers to a search object with an edge between the search object and the search word, and the search object associated with the candidate word refers to a search object with an edge between the candidate word and the search object. The two-part graph comprises a plurality of candidate words, the similarity between the search object associated with the search word and the search object associated with each candidate word is calculated, the second similarity is obtained, and after the second similarity is obtained, the first similarity is calculated.
In one embodiment, the second similarity is determined by the first similarity, and the final first similarity is obtained through multiple iterative operations. Specifically, according to the values of all edges corresponding to the search word, the values of all edges corresponding to the candidate word and the second similarity, and a preset iteration formula, carrying out iteration operation to obtain the first similarity of the search word and the candidate word. The first similarity is determined according to the values of all edges corresponding to the search word, the values of all edges corresponding to the candidate word and the second similarity, the second similarity of the two search objects is determined according to the values of all edges corresponding to each search object and the first similarity, the first similarity is determined according to the second similarity, and multiple iterations are sequentially performed to obtain the final first similarity. When the preset iteration number (for example, 20 times) is reached, the iteration is determined to be ended, and the algorithm converges to obtain the final first similarity.
By combining the values of all sides corresponding to the search word, the values of all sides corresponding to the candidate word and the similarity between the search objects to obtain the final first similarity, the search word with indirect relation and the candidate word can be established to improve the relevance of the search word and the candidate word.
In one embodiment, the formula is based on
The first similarity is calculated.
According to the formula
A second similarity is calculated.
Wherein S weighted(q,q1) represents a first similarity of the search term q to the candidate term q 1, b represents a constant, E (q) represents all search objects associated with the search term q,Ω (q, i) represents the value of the edge between the search word q and any search object i, ω (q, g) represents the value of the edge connecting the search word q and any search object g,Representing the sum of the values of all edges connecting the search terms q.
E (q 1) represents all search objects associated with search term q 1, ω (q 1, j) represents the value of the edge between the connection candidate term q 1 and any search object j, ω (q 1, h) represents the value of the edge between the connection candidate term q 1 and any search object h,/>The sum of the values representing all the edges of the join candidate word q 1.
S weighted (i, j) represents a second similarity between the search object i and the search object j, and can be obtained according to a calculation formula of the second similarity.
S weighted(a,a1) represents a second similarity of the search object a to the search object a 1, c represents a constant, E (a) represents all candidate words associated with the search object a,E (a) represents all candidate words associated with the search object a, ω (a, m) represents the value of the edge connecting the search object a and any candidate word m, ω (a, s) represents the value of the edge connecting the search object a and any candidate word s,/>Representing the sum of the values of all the edges connecting search object a.
E (a 1) represents all candidate words associated with search object a 1, ω (a 1, n) represents the value of the edge between search object a 1 and any candidate word n, ω (a 1, t) represents the value of the edge between search object a 1 and any candidate word t,/>Representing the sum of the values of all the edges connecting search object a 1.
S weighted (m, n) represents a first similarity between the candidate word m and the candidate word n, and may be obtained according to a calculation formula of the first similarity.
The electronic device can determine an initial value of the first similarity, substitutes the initial value of the first similarity into a calculation formula of the second similarity to obtain the second similarity, substitutes the obtained second similarity into the calculation formula of the first similarity to obtain the first similarity, substitutes the first similarity into the calculation formula of the second similarity to obtain the second similarity, calculates the first similarity, iterates in sequence, and obtains the final first similarity when the preset iteration times are reached.
The electronic device may also determine an initial value of the second similarity, substituting the initial value of the second similarity into a calculation formula of the first similarity to obtain the first similarity, substituting the first similarity into the calculation formula of the second similarity to obtain the second similarity, and sequentially iterating to obtain the final first similarity.
In an embodiment, according to the search object associated with the search word and the search object associated with the candidate word, a first initial similarity between the search word and the candidate word is determined, and the first initial similarity is used as an initial value of iterative operation. Illustratively, among the search objects associated with the search terms, the first initial similarity is 1 if the same search object exists, and the initial similarity is 0 if the same search object does not exist.
In an embodiment, according to the candidate word corresponding to the search object, a second initial similarity between the search object associated with the search word and the search object associated with the candidate word is determined, and the second initial similarity is used as an initial value of iterative operation. For example, among the candidate words respectively associated with any two search objects, if the same candidate word exists, the second initial similarity of the two search objects is 1, and if the same candidate word does not exist, the second initial similarity of the two search objects is 0.
In other embodiments, the first initial similarity may be set to a fixed value (e.g., 1 or 0.5), the first initial similarity may be set to an initial value of the iterative operation, or the second initial similarity may be set to a fixed value, and the second initial similarity may be set to an initial value of the iterative operation.
And calculating the similarity of the search word and the candidate word through a calculation formula of the first similarity, and fusing the ratio of the value of the edge connecting the search word and the search object to the sum of all edges connecting the search word, the ratio of the value of the edge connecting the candidate word and the search object to the sum of all edges connecting the candidate word, and the similarity between the search object connected by the search word and the search object connected by the candidate word. The ratio of the value of the edge connecting the candidate word with the search object to the sum of all edges connecting the candidate word may represent the weight each search object occupies in all search objects connecting the candidate word. And calculating the similarity between the search objects through a calculation formula of the second similarity, and fusing the ratio of the edge connecting the search objects and the candidate words to the sum of all edges connecting the search objects and the similarity between the candidate words. The ratio of the edge connecting the search object to the candidate word to the sum of all edges connecting the search object may represent the weight each candidate word occupies in all candidate words connecting the search object. And carrying out iterative operation on the first similarity and the second similarity to obtain final first similarity, and fully utilizing all edges connecting the search word and the candidate word to establish connection between the search word with an indirect relation and the candidate word, so that the correlation between the obtained first similarity and the candidate word and the correlation between the obtained first similarity and the search object are improved, and the accuracy of the subsequently determined association word is further improved.
In other embodiments, the second similarity may also be determined based on the values of all edges corresponding to the search object. For example, the second similarity may be determined based on the number of identical candidate words to which the two search objects correspond. The first similarity may also be determined according to the search object connected to the search word and the search object connected to the candidate word, for example, the first similarity may be determined according to the number of identical search objects among the search objects connected to the search word and the search objects connected to the candidate word.
S103: and determining the association word of the search word from the candidate words according to the first similarity.
Specifically, the first similarity is a first similarity of the search word with each candidate word. After the first similarity is obtained, determining the association word from the candidate words according to the first similarity corresponding to each candidate word.
In one embodiment, the candidate word with the first similarity greater than the preset value is used as the association word of the search word. The candidate words with the first similarity greater than the preset value may be candidate words with the first similarity greater than a fixed value, or may be candidate words arranged in the first N (e.g. 5) after the candidate words are ranked according to the first similarity.
It will be appreciated that the first similarity of the candidate word identical to the search word is greatest, and therefore, the associated word also includes the search word.
S104: and recommending a target search object according to the association word.
Specifically, in the bipartite graph, the associated word is one or more candidate words. And determining a search object associated with the association word according to the bipartite graph, determining a target search object from the search object, and outputting the target search object at a search interface so as to recommend the target search object to a user.
In one embodiment, a search object associated with the associated word is determined from the bipartite graph, and the search object having a correlation degree greater than a preset value is used as a target search object and output. The relevance may be determined according to one or more of search terms input by a user, search habits of the user, and search frequencies of search objects. For example, the scores of the items can be determined according to the search terms input by the user, the search habits of the user, the search frequency of the search object and a preset scoring rule, and the relevance is determined according to the scores and the corresponding weights.
In an embodiment, after determining the target search object, the target search object may be sorted in order of the higher relevance and output.
The electronic equipment determines that the search behavior starts when detecting the input behavior of the user, and determines that the search behavior ends when detecting the click behavior of the user. Characters displayed on the search interface detected between the input behavior and the click behavior are search terms of the same search behavior. The search behavior is used as a historical search behavior, the search word is used as a candidate word of the historical search behavior, the target search object clicked by the user is a search object corresponding to the historical search behavior, the historical search behavior of the bipartite graph can be updated and constructed according to the candidate word of the historical search behavior and the corresponding target search object, and then the bipartite graph is updated, so that the bipartite graph can be constructed by using richer data, and the correlation between the bipartite graph and the search behavior of the user is improved.
In an embodiment, if the candidate word same as the search word exists in the two graphs, determining a first similarity between the search word and the candidate word according to the method, determining the association word according to the first similarity, and further determining the target search object. If the candidate words which are the same as the search words do not exist in the two graphs, the target search object can be determined according to the semantics of the search words, or candidate words which are similar to the semantics of the search words or similar to the characters can be determined from the two graphs, the candidate words are used as the association words, and then the target search object is determined according to the association words.
In the above embodiment, the candidate words of the same search behavior in the two graphs include all the candidate words displayed on the display interface before the user clicks the search object, including both the incomplete word input by the user and the incorrect word input by the user. Because the candidate words of the same search behavior in the two graphs are mutually associated with the search objects corresponding to the clicking behavior, the first similarity between the search word and the candidate word is determined according to the two graphs, the associated word of more characters similar to the search word of fewer characters can be rapidly determined, or the associated word similar to the search word of wrong characters is determined, then the target object is recommended according to the associated word, the search intention of the user can be identified under the condition that fewer characters or wrong characters are input by the user, so that the input characters can be corrected, the effective search object can be recommended for the user, and the search efficiency is further improved.
The search object recommending method provided by the embodiment of the application is introduced by combining a specific scene.
As shown in fig. 5, when the mobile phone detects that the user inputs "fortune" in the case of displaying the search interface of the application program, the mobile phone determines the first similarity between "fortune" and all candidate words according to the bipartite graph, determines that the associated word of "fortune" is "sports", "sports health", "transportation", and then determines that the target search object is "sports health", "sports", "health", "step", "transportation", "driver", and then displays the target search object on the display interface according to the order of the target search object. In an embodiment, the mobile phone may also display a "more" option on the display interface, and if the user clicks the "more" option, all search objects related to the associated word are output. The mobile phone can also determine the relevance of all search objects related to the associated word, and output all search objects related to the associated word according to the sequence of the relevance from large to small.
In the above embodiment, when the user does not input the complete word, the mobile phone can determine the search intention of the user, expand the word input by the user to obtain the associated word, and further output the target search object, so that the search path can be shortened, the search efficiency can be improved, and the downloading efficiency of the application program can be further improved.
As shown in fig. 6, when the mobile phone displays a search interface of a game center, and when the mobile phone detects that a user inputs "cool", the mobile phone determines a first similarity between "cool" and all candidate words according to the bipartite graph, determines that the associated word of "cool" is "cool", "cool" according to the first similarity, and determines that the target search object is "xx cool", "xx escape", "xx fast" in order according to the associated word.
In the above embodiment, when the mobile phone detects that the user inputs the wrong character, the mobile phone can determine the association word of the user input character according to the bipartite graph, so that error correction can be performed on the character input by the user to identify the real search intention of the user, and further, a better target search object is recommended to the user, thereby improving the search efficiency.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present application.
By way of example, fig. 7 shows a schematic diagram of an architecture of the electronic device 100.
The electronic device 100 may include a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (universal serial bus, USB) interface 130, a charge management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, an earphone interface 170D, a sensor module 180, keys 190, a motor 191, an indicator 192, a camera 193, a display 194, and a subscriber identity module (subscriber identification module, SIM) card interface 195, etc. The sensor module 180 may include a pressure sensor 180A, a gyro sensor 180B, an air pressure sensor 180C, a magnetic sensor 180D, an acceleration sensor 180E, a distance sensor 180F, a proximity sensor 180G, a fingerprint sensor 180H, a temperature sensor 180J, a touch sensor 180K, an ambient light sensor 180L, a bone conduction sensor 180M, and the like.
It should be understood that the illustrated structure of the embodiment of the present application does not constitute a specific limitation on the electronic device 100. In other embodiments of the application, electronic device 100 may include more or fewer components than shown, or certain components may be combined, or certain components may be split, or different arrangements of components. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.
The processor 110 may include one or more processing units, such as: the processor 110 may include an application processor (application processor, AP), a modem processor, a graphics processor (graphics processing unit, GPU), an image signal processor (IMAGE SIGNAL processor, ISP), a controller, a video codec, a digital signal processor (DIGITAL SIGNAL processor, DSP), a baseband processor, and/or a neural-Network Processor (NPU), etc. Wherein the different processing units may be separate devices or may be integrated in one or more processors.
The controller can generate operation control signals according to the instruction operation codes and the time sequence signals to finish the control of instruction fetching and instruction execution.
A memory may also be provided in the processor 110 for storing instructions and data. In some embodiments, the memory in the processor 110 is a cache memory. The memory may hold instructions or data that the processor 110 has just used or recycled. If the processor 110 needs to reuse the instruction or data, it can be called directly from the memory. Repeated accesses are avoided and the latency of the processor 110 is reduced, thereby improving the efficiency of the system.
In some embodiments, the processor 110 may include one or more interfaces. The interfaces may include an integrated circuit (inter-INTEGRATED CIRCUIT, I2C) interface, an integrated circuit built-in audio (inter-INTEGRATED CIRCUIT SOUND, I2S) interface, a pulse code modulation (pulse code modulation, PCM) interface, a universal asynchronous receiver transmitter (universal asynchronous receiver/transmitter, UART) interface, a mobile industry processor interface (mobile industry processor interface, MIPI), a general-purpose input/output (GPIO) interface, a subscriber identity module (subscriber identity module, SIM) interface, and/or a universal serial bus (universal serial bus, USB) interface, among others.
It should be understood that the interfacing relationship between the modules illustrated in the embodiments of the present application is only illustrative, and is not meant to limit the structure of the electronic device 100. In other embodiments of the present application, the electronic device 100 may also employ different interfacing manners in the above embodiments, or a combination of multiple interfacing manners.
The wireless communication function of the electronic device 100 may be implemented by the antenna 1, the antenna 2, the mobile communication module 150, the wireless communication module 160, a modem processor, a baseband processor, and the like.
The electronic device 100 implements display functions through a GPU, a display screen 194, an application processor, and the like. The GPU is a microprocessor for image processing, and is connected to the display 194 and the application processor. The GPU is used to perform mathematical and geometric calculations for graphics rendering. Processor 110 may include one or more GPUs that execute program instructions to generate or change display information.
The display screen 194 is used to display images, videos, and the like. The display 194 includes a display panel. The display panel may employ a Liquid Crystal Display (LCD) CRYSTAL DISPLAY, an organic light-emitting diode (OLED), an active-matrix organic LIGHT EMITTING diode (AMOLED), a flexible light-emitting diode (FLED), miniled, microLed, micro-oLed, a quantum dot LIGHT EMITTING diode (QLED), or the like. In some embodiments, the electronic device 100 may include 1 or N display screens 194, N being a positive integer greater than 1.
The electronic device 100 may implement photographing functions through an ISP, a camera 193, a video codec, a GPU, a display screen 194, an application processor, and the like.
The external memory interface 120 may be used to connect an external memory card, such as a Micro SD card, to enable expansion of the memory capabilities of the electronic device 100. The external memory card communicates with the processor 110 through an external memory interface 120 to implement data storage functions. For example, files such as music, video, etc. are stored in an external memory card.
The internal memory 121 may be used to store computer executable program code including instructions. The internal memory 121 may include a storage program area and a storage data area. The storage program area may store an application program (such as a sound playing function, an image playing function, etc.) required for at least one function of the operating system, etc. The storage data area may store data created during use of the electronic device 100 (e.g., audio data, phonebook, etc.), and so on. In addition, the internal memory 121 may include a high-speed random access memory, and may further include a nonvolatile memory such as at least one magnetic disk storage device, a flash memory device, a universal flash memory (universal flash storage, UFS), and the like. The processor 110 performs various functional applications of the electronic device 100 and data processing by executing instructions stored in the internal memory 121 and/or instructions stored in a memory provided in the processor.
The electronic device 100 may implement audio functions through an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, an earphone interface 170D, an application processor, and the like. Such as music playing, recording, etc.
The touch sensor 180K, also referred to as a "touch device". The touch sensor 180K may be disposed on the display screen 194, and the touch sensor 180K and the display screen 194 form a touch screen, which is also called a "touch screen". The touch sensor 180K is for detecting a touch operation acting thereon or thereabout. The touch sensor may communicate the detected touch operation to the application processor to determine the touch event type. Visual output related to touch operations may be provided through the display 194. In other embodiments, the touch sensor 180K may also be disposed on the surface of the electronic device 100 at a different location than the display 194.
The software system of the electronic device 100 may employ a layered architecture, an event driven architecture, a microkernel architecture, a microservice architecture, or a cloud architecture. In the embodiment of the invention, taking an Android system with a layered architecture as an example, a software structure of the electronic device 100 is illustrated.
Fig. 8 is a software configuration block diagram of the electronic device 100 according to the embodiment of the present invention.
The layered architecture divides the software into several layers, each with distinct roles and branches. The layers communicate with each other through a software interface. In some embodiments, the Android system is divided into four layers, from top to bottom, an application layer, an application framework layer, an Zhuoyun rows (Android runtime) and system libraries, and a kernel layer, respectively.
The application layer may include a series of application packages.
As shown in fig. 8, the application package may include applications for cameras, gallery, calendar, phone calls, maps, navigation, WLAN, bluetooth, music, video, short messages, etc.
The application framework layer provides an application programming interface (application programming interface, API) and programming framework for the application of the application layer. The application framework layer includes a number of predefined functions.
As shown in fig. 8, the application framework layer may include a window manager, a content provider, a view system, a phone manager, a resource manager, a notification manager, and the like.
The window manager is used for managing window programs. The window manager can acquire the size of the display screen, judge whether a status bar exists, lock the screen, intercept the screen and the like.
The content provider is used to store and retrieve data and make such data accessible to applications. The data may include video, images, audio, calls made and received, browsing history and bookmarks, phonebooks, etc.
The view system includes visual controls, such as controls to display text, controls to display pictures, and the like. The view system may be used to build applications. The display interface may be composed of one or more views. For example, a display interface including a text message notification icon may include a view displaying text and a view displaying a picture.
The telephony manager is used to provide the communication functions of the electronic device 100. Such as the management of call status (including on, hung-up, etc.).
The resource manager provides various resources for the application program, such as localization strings, icons, pictures, layout files, video files, and the like.
The notification manager allows the application to display notification information in a status bar, can be used to communicate notification type messages, can automatically disappear after a short dwell, and does not require user interaction. Such as notification manager is used to inform that the download is complete, message alerts, etc. The notification manager may also be a notification in the form of a chart or scroll bar text that appears on the system top status bar, such as a notification of a background running application, or a notification that appears on the screen in the form of a dialog window. For example, a text message is prompted in a status bar, a prompt tone is emitted, the electronic device vibrates, and an indicator light blinks, etc.
Android run time includes a core library and virtual machines. Android runtime is responsible for scheduling and management of the android system.
The core library consists of two parts: one part is a function which needs to be called by java language, and the other part is a core library of android.
The application layer and the application framework layer run in a virtual machine. The virtual machine executes java files of the application program layer and the application program framework layer as binary files. The virtual machine is used for executing the functions of object life cycle management, stack management, thread management, security and exception management, garbage collection and the like.
The system library may include a plurality of functional modules. For example: surface manager (surface manager), media Libraries (Media Libraries), three-dimensional graphics processing Libraries (e.g., openGL ES), 2D graphics engines (e.g., SGL), etc.
The surface manager is used to manage the display subsystem and provides a fusion of 2D and 3D layers for multiple applications.
Media libraries support a variety of commonly used audio, video format playback and recording, still image files, and the like. The media library may support a variety of audio and video encoding formats, such as MPEG4, h.264, MP3, AAC, AMR, JPG, PNG, etc.
The three-dimensional graphic processing library is used for realizing three-dimensional graphic drawing, image rendering, synthesis, layer processing and the like.
The 2D graphics engine is a drawing engine for 2D drawing.
The kernel layer is a layer between hardware and software. The inner core layer at least comprises a display driver, a camera driver, an audio driver and a sensor driver.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and in part, not described or illustrated in any particular embodiment, reference is made to the related descriptions of other embodiments.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions. The functional units and modules in the embodiment may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit, where the integrated units may be implemented in a form of hardware or a form of a software functional unit. In addition, the specific names of the functional units and modules are only for distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working process of the units and modules in the above system may refer to the corresponding process in the foregoing method embodiment, which is not described herein again.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the present application may implement all or part of the flow of the method of the above embodiments, and may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of each of the method embodiments described above. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include at least: any entity or device capable of carrying computer program code to a camera device/electronic apparatus, a recording medium, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, and a software distribution medium. Such as a U-disk, removable hard disk, magnetic or optical disk, etc.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus/device and method may be implemented in other manners. For example, the apparatus/device embodiments described above are merely illustrative, e.g., the division of the modules or units is merely a logical functional division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via interfaces, devices or units, which may be in electrical, mechanical or other forms.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
Finally, it should be noted that: the foregoing is merely illustrative of specific embodiments of the present application, and the scope of the present application is not limited thereto, but any changes or substitutions within the technical scope of the present application should be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (8)

1. A search object recommendation method, comprising:
Acquiring search words;
Determining a first similarity between the search word and a candidate word according to a bipartite graph, wherein the bipartite graph is generated according to historical search behaviors, the bipartite graph comprises a first set, a second set and a plurality of edges, the first set comprises the candidate word, the second set comprises a search object, each edge is used for connecting the candidate word and the search object which are mutually related, and the value of each edge represents the degree of association between the candidate word and the search object which are mutually related; the candidate words of the same search behavior in the historical search behaviors are mutually associated with the search objects corresponding to the click behaviors, and the candidate words of the same search behavior represent all candidate words detected from the detection of the input behavior to the detection of the click behavior; the determining the first similarity between the search word and the candidate word according to the bipartite graph comprises the following steps: performing iterative operation according to the values of all edges corresponding to the search word, the values of all edges corresponding to the candidate word, a second similarity and a preset iterative formula to obtain a first similarity between the search word and the candidate word, wherein the second similarity is the similarity between a search object associated with the search word and a search object associated with the candidate word, and the second similarity is determined by the first similarity;
Determining association words of the search words from the candidate words according to the first similarity;
and recommending a target search object according to the association word.
2. The method of claim 1, wherein the value of the edge is determined based on a number of clicks of the candidate word, the number of clicks of the candidate word representing a number of clicks of the corresponding search object based on the candidate word.
3. The method of claim 1, wherein before performing the iterative operation according to the values of all edges corresponding to the search term, the values of all edges corresponding to the candidate term, the second similarity, and a preset iterative formula, the method further comprises:
Determining a first initial similarity between the search word and the candidate word according to the search object associated with the search word and the search object associated with the candidate word, taking the first initial similarity as an initial value of the iterative operation, or,
And determining a second initial similarity between the search object associated with the search word and the search object associated with the candidate word according to the candidate word corresponding to the search object, and taking the second initial similarity as an initial value of the iterative operation.
4. The method of claim 1, wherein the determining the associated word of the search word from the candidate words based on the first similarity comprises:
and taking the candidate words with the first similarity larger than a preset value as the association words of the search words.
5. The method of claim 1, wherein the number of the associated words is a plurality, and the recommending the target search object according to the associated words comprises:
Determining the search object associated with each associated word;
and taking the search object with the relevance being larger than a preset value as the target search object and outputting the target search object.
6. The method of claim 1, wherein after recommending a target search object according to the associative word, the method further comprises:
And updating the bipartite graph according to the target search object clicked by the user.
7. An electronic device comprising a processor for executing a computer program stored in a memory to implement the method of any one of claims 1 to 6.
8. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the method according to any one of claims 1 to 6.
CN202310143444.XA 2023-02-09 2023-02-09 Search object recommendation method, electronic device and computer readable storage medium Active CN117131241B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310143444.XA CN117131241B (en) 2023-02-09 2023-02-09 Search object recommendation method, electronic device and computer readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310143444.XA CN117131241B (en) 2023-02-09 2023-02-09 Search object recommendation method, electronic device and computer readable storage medium

Publications (2)

Publication Number Publication Date
CN117131241A CN117131241A (en) 2023-11-28
CN117131241B true CN117131241B (en) 2024-05-24

Family

ID=88849691

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310143444.XA Active CN117131241B (en) 2023-02-09 2023-02-09 Search object recommendation method, electronic device and computer readable storage medium

Country Status (1)

Country Link
CN (1) CN117131241B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103092828A (en) * 2013-02-06 2013-05-08 杭州电子科技大学 Text similarity measuring method based on semantic analysis and semantic relation network
CN103473283A (en) * 2013-08-29 2013-12-25 中国测绘科学研究院 Method for matching textual cases
CN104657514A (en) * 2015-03-24 2015-05-27 成都知数科技有限公司 Synonym identification method based on electronic commerce user behavior data
CN104915860A (en) * 2015-06-10 2015-09-16 无线生活（杭州）信息科技有限公司 Commodity recommendation method and device
CN106445989A (en) * 2016-06-03 2017-02-22 新乡学院 Query click graph-based search recommendation model optimization
CN108227955A (en) * 2017-12-29 2018-06-29 北京奇虎科技有限公司 It is a kind of that the method and device for recommending input association is searched for based on user's history
CN111368038A (en) * 2020-03-09 2020-07-03 广州市百果园信息技术有限公司 Keyword extraction method and device, computer equipment and storage medium
CN111488434A (en) * 2019-01-25 2020-08-04 北京字节跳动网络技术有限公司 Recommendation method and device for input associative words, storage medium and electronic equipment
CN114064846A (en) * 2020-07-31 2022-02-18 中移(苏州)软件技术有限公司 Theme similarity determination method and device, electronic equipment and storage medium

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4961755B2 (en) * 2006-01-23 2012-06-27 富士ゼロックス株式会社 Word alignment device, word alignment method, word alignment program
US11874882B2 (en) * 2019-07-02 2024-01-16 Microsoft Technology Licensing, Llc Extracting key phrase candidates from documents and producing topical authority ranking

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103092828A (en) * 2013-02-06 2013-05-08 杭州电子科技大学 Text similarity measuring method based on semantic analysis and semantic relation network
CN103473283A (en) * 2013-08-29 2013-12-25 中国测绘科学研究院 Method for matching textual cases
CN104657514A (en) * 2015-03-24 2015-05-27 成都知数科技有限公司 Synonym identification method based on electronic commerce user behavior data
CN104915860A (en) * 2015-06-10 2015-09-16 无线生活（杭州）信息科技有限公司 Commodity recommendation method and device
CN106445989A (en) * 2016-06-03 2017-02-22 新乡学院 Query click graph-based search recommendation model optimization
CN108227955A (en) * 2017-12-29 2018-06-29 北京奇虎科技有限公司 It is a kind of that the method and device for recommending input association is searched for based on user's history
CN111488434A (en) * 2019-01-25 2020-08-04 北京字节跳动网络技术有限公司 Recommendation method and device for input associative words, storage medium and electronic equipment
CN111368038A (en) * 2020-03-09 2020-07-03 广州市百果园信息技术有限公司 Keyword extraction method and device, computer equipment and storage medium
CN114064846A (en) * 2020-07-31 2022-02-18 中移(苏州)软件技术有限公司 Theme similarity determination method and device, electronic equipment and storage medium

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Renchi Yang.Efficient and Effective Similarity Search over Bipartite Graphs.《WWW '22: Proceedings of the ACM Web Conference 2022》.2022,308–318. *
基于二部图的推荐算法的研究与应用;陈诚;《中国优秀硕士学位论文全文数据库 信息科技辑》;20180415(第04期);I138-3793 *
基于位置的个性化关键词查询推荐;梁耀培 等;《深圳大学学报(理工版)》;20190730;第36卷(第04期);467-472 *

Also Published As

Publication number Publication date
CN117131241A (en) 2023-11-28

Similar Documents

Publication Publication Date Title
CN112445922B (en) Picture processing method and device
US11017045B2 (en) Personalized user experience and search-based recommendations
US8275399B2 (en) Dynamic context-data tag cloud
EP2895967A1 (en) Gesture-based search queries
US11604819B2 (en) Associating a graphical element to media content item collections
WO2021147421A1 (en) Automatic question answering method and apparatus for man-machine interaction, and intelligent device
US20120079009A1 (en) Method and apparatus for choosing resources based on context and inheritance
US10540415B2 (en) Apparatus and method for managing history information in an electronic device
CN112269853A (en) Search processing method, search processing device and storage medium
CN111966491A (en) Method for counting occupied memory and terminal equipment
CN111966894A (en) Information query method and device, storage medium and electronic equipment
CN117131241B (en) Search object recommendation method, electronic device and computer readable storage medium
CN116028147B (en) Application program recommendation method and electronic equipment
US20240248901A1 (en) Method and system of using domain specific knowledge in retrieving multimodal assets
CN117707716A (en) Thread scheduling method, electronic device and computer readable storage medium
CN116958715A (en) Method and device for detecting hand key points and storage medium
CN113641292B (en) Method and electronic equipment for operating on touch screen
CN116071614A (en) Sample data processing method, related device and storage medium
CN109901859A (en) Dynamic configuration official documents and correspondence method, electronic device and storage medium
WO2021146941A1 (en) Disease location acquisition method, apparatus, device and computer readable storage medium
CN113822011A (en) Font switching method and electronic equipment
CN117724780B (en) Information acquisition method
WO2024131584A1 (en) Trajectory playback method and apparatus
CN114546113B (en) Menu operation method and device, storage medium and electronic equipment
CN111259161B (en) Ontology establishing method and device and storage medium

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