CN112148960B - Method, device, equipment and storage medium for determining category of attention point - Google Patents
Method, device, equipment and storage medium for determining category of attention point Download PDFInfo
- Publication number
- CN112148960B CN112148960B CN201910572735.4A CN201910572735A CN112148960B CN 112148960 B CN112148960 B CN 112148960B CN 201910572735 A CN201910572735 A CN 201910572735A CN 112148960 B CN112148960 B CN 112148960B
- Authority
- CN
- China
- Prior art keywords
- preset
- probability
- point
- category
- attention point
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 8
- 239000000758 substrate Substances 0.000 claims 1
- 241000220225 Malus Species 0.000 description 49
- 230000006399 behavior Effects 0.000 description 45
- 230000015654 memory Effects 0.000 description 21
- 235000013399 edible fruits Nutrition 0.000 description 15
- 230000006870 function Effects 0.000 description 15
- 238000004891 communication Methods 0.000 description 9
- 235000021016 apples Nutrition 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000003491 array Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
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)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention provides a category determination method, a device, equipment and a storage medium of a focus point, wherein the method comprises the following steps: acquiring a preset attention point of a single user, and acquiring each associated attention point of the preset attention point; determining the individuation probability that the preset attention point belongs to the preset category according to the global probability that the preset attention point belongs to the preset category, the global probability that each associated attention point belongs to the preset category and the association degree of the preset attention point and each associated attention point; wherein the global probability is: in the behavior data of the global user, the probability that the preset attention point or the related attention point belongs to the preset category is determined; the individualization probability is as follows: in the behavior data of the single user, the probability that the predetermined point of interest belongs to the predetermined category. The method and the device can determine the probability that the attention point belongs to different categories for a single user.
Description
Technical Field
The present invention relates to the field of personalized recommendation technologies, and in particular, to a method, an apparatus, a device, and a storage medium for determining a category of a focus.
Background
The focus of the user is typically expressed as a single word. When personalized recommendation is carried out for the user, the interest of the user can be known through the attention point of the user, and the content conforming to the interest of the user is recommended for the user. However, there are ambiguities in expressing the point of interest in a single word. For example, for the point of interest "apple", it is not possible to determine whether the point of interest refers to a digital product of the relevant brand or to fruit. If the category to which the point of interest belongs can be annotated, the ambiguity can be largely resolved.
Currently, a method for determining the category of the attention point of the global user appears, for example, according to behavior data such as searching behavior of the global user, the probability that the attention point "apple" belongs to the "digital" category is determined to be 0.8, and the probability that the attention point "apple" belongs to the "fruit" category is determined to be 0.2. The class probability of the point of interest "apple" can be expressed as (apple- > number, 0.8), (apple- > fruit, 0.2).
However, the aforementioned point of interest category probabilities are for global users, not for individual users. There is a high probability that a single user's point of interest belongs to a category and there is a bias between the probability that the same point of interest of the global user belongs to that category. For example, it is likely that the probability of "apples" of the "fruit" category being of interest is greater for fruit farmers than for "apples" of the "digital" category. The prior art fails to determine the probability that a point of interest belongs to different categories for a single user.
Disclosure of Invention
The embodiment of the invention provides a category determination method and device of a focus point, which are used for at least solving the technical problems in the prior art.
In a first aspect, an embodiment of the present invention provides a method for determining a category of a point of interest, including:
acquiring a preset attention point of a single user, and acquiring each associated attention point of the preset attention point;
determining the individuation probability that the preset attention point belongs to the preset category according to the global probability that the preset attention point belongs to the preset category, the global probability that each associated attention point belongs to the preset category and the association degree of the preset attention point and each associated attention point; wherein,
the global probability is: in the behavior data of the global user, the probability that the preset attention point or the related attention point belongs to the preset category is determined;
the individualization probability is as follows: in the behavior data of the single user, the probability that the predetermined point of interest belongs to the predetermined category.
In one embodiment, the global probability is obtained from a global point of interest network;
the global point of interest network comprises: in the behavior data of the global user, probabilities that all the concerned points belong to different categories are provided.
In one embodiment, before determining the personalized probability that the predetermined point of interest belongs to the predetermined category, the method further includes:
according to the behavior data of the single user, determining a directed edge formed by N attention points contained in each behavior data; the N is an integer greater than 1; assigning the weight of the directed edge as the number of times that the directed edge appears in each behavior data;
according to any two adjacent attention points in each directed edge, determining an undirected edge formed by any two adjacent attention points;
assigning the weight of the undirected edge as the sum of the weights of the related directed edges; wherein the related directed edge is a directed edge comprising the arbitrary two adjacent points of interest.
In one embodiment, the degree of association is: and the weight of the undirected edge formed by the preset attention point and the associated attention point.
In one embodiment, the determining the personalized probability that the predetermined attention point belongs to the predetermined category according to the global probability that the predetermined attention point belongs to the predetermined category, the global probability that the respective associated attention point belongs to the predetermined category, and the association degree of the predetermined attention point and the respective associated attention point includes:
the following equation is used for determination:
wherein,
the P' (X, S) is the individualization probability that the preset attention point X belongs to the preset category S;
the P (X, S) is the global probability that the predetermined focus X belongs to the predetermined category S;
said P (Y) i S) is the point of interest Y i Global probabilities belonging to a predetermined category S; the point of interest Y i An associated point of interest for the predetermined point of interest X;
the W is i For the focus X and the focus Y i The weight of the formed undirected edge;
the N is the number of associated attention points of the preset attention point X;
the a and the b are preset parameters.
In one embodiment, the method further comprises:
determining the individuation probability that the preset attention point belongs to other categories;
according to the personalized probability that the preset attention point belongs to the preset category and the personalized probability that the preset attention point belongs to other categories, normalized exponential function calculation is carried out on the personalized probability that the preset attention point belongs to the preset category, and the value of the personalized probability that the preset attention point belongs to the preset category is modified to be the result of the calculation.
In one embodiment, after determining the directed edges formed by the N points of interest included in each behavior data, the method further includes: determining a time stamp of the directed edge according to the last update time of the behavior data corresponding to the directed edge;
the method further comprises the steps of: and predicting the user behavior of the single user according to the directed edge and the time stamp of the directed edge.
In a second aspect, an embodiment of the present invention further proposes a category determining apparatus of a point of interest, including:
the acquisition module is used for acquiring the preset attention point of a single user and acquiring each associated attention point of the preset attention point;
the determining module is used for determining the individuation probability that the preset attention point belongs to the preset category according to the global probability that the preset attention point belongs to the preset category, the global probability that each associated attention point belongs to the preset category and the association degree of the preset attention point and each associated attention point; wherein the global probability is: in the behavior data of the global user, the probability that the preset attention point or the related attention point belongs to the preset category is determined; the individualization probability is as follows: in the behavior data of the single user, the probability that the predetermined point of interest belongs to the predetermined category.
In one embodiment, the determining module obtains the global probability from a global point of interest network;
the global point of interest network comprises: in the behavior data of the global user, probabilities that all the concerned points belong to different categories are provided.
In one embodiment, the apparatus further comprises:
the personalized attention point network determining module is used for determining a directed edge formed by N attention points contained in each behavior data according to the behavior data of the single user; the N is an integer greater than 1; assigning the weight of the directed edge as the number of times that the directed edge appears in each behavior data; according to any two adjacent attention points in each directed edge, determining an undirected edge formed by any two adjacent attention points; assigning the weight of the undirected edge as the sum of the weights of the related directed edges; wherein the related directed edge is a directed edge comprising the arbitrary two adjacent points of interest.
In one embodiment, the degree of association is: and the weight of the undirected edge formed by the preset attention point and the associated attention point.
In one embodiment, the determining module is configured to determine the personalized probability that the predetermined point of interest belongs to the predetermined category using the following equation:
wherein,
the P' (X, S) is the individualization probability that the preset attention point X belongs to the preset category S;
the P (X, S) is the global probability that the predetermined focus X belongs to the predetermined category S;
said P (Y) i S) is the point of interest Y i Global probabilities belonging to a predetermined category S; the point of interest Y i An associated point of interest for the predetermined point of interest X;
the W is i For the focus X and the focus Y i The weight of the formed undirected edge;
the N is the number of associated attention points of the preset attention point X;
the a and the b are preset parameters.
In one embodiment, the determining module is further configured to: determining the individuation probability that the preset attention point belongs to other categories; according to the personalized probability that the preset attention point belongs to the preset category and the personalized probability that the preset attention point belongs to other categories, normalized exponential function calculation is carried out on the personalized probability that the preset attention point belongs to the preset category, and the value of the personalized probability that the preset attention point belongs to the preset category is modified to be the result of the calculation.
In one embodiment, the personalized focus network determining module is further configured to determine a timestamp of the directed edge according to a last update time of the behavior data corresponding to the directed edge;
the apparatus further comprises: and the prediction module is used for predicting the user behavior of the single user according to the directed edge and the time stamp of the directed edge.
In a third aspect, an embodiment of the present invention provides a category determining device for a point of interest, where a function of the device may be implemented by hardware, or may be implemented by executing corresponding software by hardware. The hardware or software includes one or more modules corresponding to the functions described above.
In one possible design, the structure of the device includes a processor and a memory, where the memory is configured to store a program for supporting the device to perform the category determination method of the point of interest described above, and the processor is configured to execute the program stored in the memory. The device may also include a communication interface for communicating with other devices or communication networks.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing computer software instructions for use by a category determining device of a point of interest, including a program for executing the category determining method of a point of interest described above.
In a fifth aspect, embodiments of the present invention provide a computer program product comprising a computer program which, when executed by a processor, implements a method as described above.
One of the above technical solutions has the following advantages or beneficial effects:
the category determining method of the attention points provided by the embodiment of the invention can determine the probability that the attention points of the single user belong to different categories according to the probability that the attention points of the global user belong to different categories and the related information of the preset attention points and the related attention points of the single user.
The foregoing summary is for the purpose of the specification only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features of the present invention will become apparent by reference to the drawings and the following detailed description.
Drawings
In the drawings, the same reference numerals refer to the same or similar parts or elements throughout the several views unless otherwise specified. The figures are not necessarily drawn to scale. It is appreciated that these drawings depict only some embodiments according to the disclosure and are not therefore to be considered limiting of its scope.
FIG. 1 is a flowchart illustrating a method for determining a category of a point of interest according to an embodiment of the present invention;
FIG. 2 is a second flowchart of a method for determining a category of a focus according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a class determination device according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a category determining device with a focus according to a second embodiment of the present invention;
fig. 5 is a schematic structural diagram of a category determining device of an attention point according to an embodiment of the present invention.
Detailed Description
Hereinafter, only certain exemplary embodiments are briefly described. As will be recognized by those of skill in the pertinent art, the described embodiments may be modified in various different ways without departing from the spirit or scope of the present invention. Accordingly, the drawings and description are to be regarded as illustrative in nature and not as restrictive.
The embodiment of the invention mainly provides a method and a device for determining the category of a focus, and the following description of the technical scheme is respectively carried out through the following embodiments.
Fig. 1 is a flowchart of a method for determining a category of a point of interest according to an embodiment of the present invention, including:
s11: acquiring a preset attention point of a single user, and acquiring each associated attention point of the preset attention point;
s12: determining the individuation probability that the preset attention point belongs to the preset category according to the global probability that the preset attention point belongs to the preset category, the global probability that each associated attention point belongs to the preset category and the association degree of the preset attention point and each associated attention point; wherein,
the global probability is: in the behavior data of the global user, the probability that the preset attention point or the related attention point belongs to the preset category is determined;
the individualization probability is as follows: in the behavior data of the single user, the probability that the predetermined point of interest belongs to the predetermined category.
In one possible implementation, the global probability is obtained from a global point of interest network;
the global point of interest network comprises: in the behavior data of the global user, probabilities that all the concerned points belong to different categories are provided.
The global point of interest network can be represented by a G-Net.
The behavior data may include search behavior, browsing behavior, and the like.
The probability that the point of interest belongs to different categories may be expressed as (point of interest- > category, probability). For example, the point of interest "apple" may belong to both the "digital" and "fruit" categories, where the probability of belonging to the "digital" category is 0.8 and the probability of belonging to the "fruit" category is 0.2. For the point of interest "apple", the probabilities that it belongs to different categories can be expressed in the following form:
(apple- > number, 0.8);
(apple— fruit, 0.2);
the sum of the two probabilities is equal to 1.
For global users, the probabilities that the various points of interest belong to different categories form a global point of interest network, namely G-Net. In addition, the G-Net can also include co-occurrence relationships between points of interest and points of interest. For example, (points of interest 1, …, points of interest m), m >1, represent that points of interest 1, …, point of interest m, appear in the same piece of behavior data.
In one possible implementation, G-Net, and a personalized point of interest network (hereinafter U-Net) for an individual user may be employed to determine a personalized probability that the individual user's point of interest belongs to a predetermined category. Fig. 2 is a flowchart second implementation of a method for determining a category of a focus point according to an embodiment of the present invention, and the step S12 further includes:
s21: according to the behavior data of the single user, determining a directed edge formed by N attention points contained in each behavior data; the N is an integer greater than 1; assigning the weight of the directed edge as the number of times that the directed edge appears in each behavior data;
s22: according to any two adjacent attention points in each directed edge, determining an undirected edge formed by any two adjacent attention points;
s23: assigning the weight of the undirected edge as the sum of the weights of the related directed edges; wherein the related directed edge is a directed edge comprising the arbitrary two adjacent points of interest.
The directional edges and the undirected edges determined in steps S21 to S23 constitute the above-mentioned U-Net.
For example, the user a has 4 pieces of behavior data at times t1, t2, t3, and t4, where t1, t2, t3, and t4 are in the order from front to back.
According to the behavior data at the time t1, determining a directed edge formed by the 3 included attention points, namely, a directed edge (apple, mobile phone, iPhoneX), assigning an ID (for example, a hash ID) to the directed edge, and assigning the weight of the directed edge as 1. The embodiment of the invention can set the time stamp of the directed edge to t1. And according to the directed edge, the undirected edge formed by any two adjacent attention points, namely (apple, mobile phone), (mobile phone, iPhoneX) can be determined. The weights of the undirected edge (apple, mobile phone) and the undirected edge (mobile phone, iPhoneX) are respectively assigned to be 1.
According to the behavior data at the time t2, determining a directed edge formed by the 3 included attention points, namely, a directed edge (apple, mobile phone, iPhone 7), assigning an ID (for example, a hash ID) to the directed edge, and assigning the weight of the directed edge as 1. The embodiment of the invention can set the time stamp of the directed edge to t2. And according to the directed edge, the undirected edge formed by any two adjacent attention points, namely (apple, mobile phone), (mobile phone, iPhone 7) can be determined. Since the undirected edge (apple, mobile phone) appears before, the weight of the undirected edge (apple, mobile phone) is added with 1, and the weight of the undirected edge (apple, mobile phone) is changed to 2. The weight of the undirected edge (phone 7) is assigned to 1.
According to the behavior data at the time t3, determining a directed edge (apple, price) formed by the contained 2 attention points, assigning an ID (e.g. hash ID) to the directed edge, and assigning the weight of the directed edge as 1. The embodiment of the invention can set the time stamp of the directed edge to t3. And according to the directed edge, the undirected edge formed by two adjacent attention points, namely, the undirected edge (apple, price) can be determined, and the weight of the undirected edge (apple, price) is assigned to be 1.
From the behavior data at time t4, a directed edge composed of 2 points of interest included, i.e., a directed edge (apple, price) is determined, since this directed edge has been generated at time t3 and assigned a fixed ID. Thus, the timestamp of the directed edge may be modified to t4, thereby indicating the last update time of the directed edge; and adds 1 to the weight of the directed edge. In this embodiment, the weight of the directed edge (apple, price) is changed to 2. The weight of the undirected edge (apple, price) may be added by 1.
It can be seen from the above procedure that the final calculation result of the weights of the undirected edges is equal to the sum of the weights of the related directed edges.
In one possible implementation, the timestamp of the undirected edge may also be determined from the timestamp of the associated directed edge. For example, in the above example, the relevant directed edges of the undirected edge (apple, cell phone) include: directed edges (apple, cell phone, iPhoneX) and directed edges (apple, cell phone, iPhone 7), the timestamp of the undirected edge (apple, cell phone) is determined to be equal to the timestamp of the directed edge (apple, cell phone, iPhone 7), i.e. equal to t2. As another example, in the above example, the relevant directed edge of the undirected edge (apple, price) is the directed edge (apple, price), then the timestamp of the undirected edge (apple, price) is determined to be equal to the timestamp of the directed edge (apple, price), i.e. equal to t4.
In one possible implementation manner, the weight of the undirected edge reflects the association degree of two concerns contained in the undirected edge. The larger the weight value is, the greater the association degree of the two concerns contained in the weight value is.
In one possible implementation, with the G-Net and U-Net described above, step S12 includes:
determining a personalized probability that a predetermined point of interest belongs to the predetermined category using equation (1):
wherein, the P' (X, S) is a personalized probability that the predetermined focus X belongs to a predetermined category S;
the P (X, S) is the global probability that the predetermined focus X belongs to the predetermined category S;
said P (Y) i S) is the point of interest Y i Global probabilities belonging to a predetermined category S; the point of interest Y i An associated point of interest for the predetermined point of interest X;
the W is i For the focus X and the focus Y i The weight of the formed undirected edge;
the N is the number of associated attention points of the preset attention point X;
the a and the b are preset parameters.
In addition, the calculation result may also be calculated by a normalized exponential function (i.e. Softmax function), which specifically includes:
determining the individuation probability that the preset attention point belongs to other categories;
according to the personalized probability that the preset attention point belongs to the preset category and the personalized probability that the preset attention point belongs to other categories, normalized exponential function calculation is carried out on the personalized probability that the preset attention point belongs to the preset category, and the value of the personalized probability that the preset attention point belongs to the preset category is modified to be the result of the calculation.
For example, U-Net for user A includes:
a directed edge (apple, mobile phone, iPhoneX), whose ID is hash1;
a directed edge (apple, mobile phone, iPhone 7), whose ID is hash2;
directed edges (apples, prices) with an ID of hash3;
undirected edges (apples, mobile phones), weight is 2;
undirected edges (apples, prices), weight 1;
undirected edge (iphonex), weight 1;
undirected edge (phone 7), weight 1.
In the G-Net of the global user, the class probabilities of the relevant focuses include:
(apple- > number, 0.8);
(apple— fruit, 0.2);
(handset- > number, 1.0);
(price- > number, 0.05);
(price— fruit, 0.05);
it is assumed that in the above formula (1), a=0.4 and b=0.6. Based on the above, the personalized probability that the focus "apple" belongs to the "digital" category can be calculated as follows:
the personalized probabilities that the focus "apple" belongs to the "fruit" category are:
the user's focus "apple" may belong to the two categories, and according to the settlement result, softmax function calculation may be performed on P ' (apple, digital) and P ' (apple, fruit), and the values of P ' (apple, digital) and P ' (apple, fruit) are modified to be the calculated result. In particular, the method comprises the steps of,
after calculation, the sum of the final P '(apple, fruit) and P' (apple, digital) is equal to 1.
The personalized probability that the preset attention points belong to the preset categories can be adopted to infer or predict the behaviors of the users. In a possible implementation manner, after the directional edge is determined in the step S21, the method may further include: determining a time stamp of the directed edge according to the last update time of the behavior data corresponding to the directed edge;
the method may further include: and predicting the user behavior of the single user according to the directed edges and the time stamps of the directed edges.
For example, at time t5, user A pays attention (apple, coupon), assuming that the class probability of the focus "coupon" in G-Net is: (preferential number, 0.05), (preferential fruit, 0.05). According to the category probability U-Net, preference information of apples of the digital category which is more focused by the user A than other users can be predicted.
As another example, based on the timestamp t1 of the directed edge (apple, cell phone, iPhone x) of user a, the timestamp t2 of the directed edge (apple, cell phone, iPhone 7), and the timing of t1 and t2, it can be predicted that user a is more concerned with cheaper models than iPhone x, such as iPhone7 or models between iPhone7 and iPhone x.
Therefore, the category determination method of the attention points provided by the embodiment of the invention can determine the probability that each attention point belongs to different categories for a single user, so that the individuation of the user is realized, and the content represented by the user paying attention to something is more accurate. By building a personalized point of interest network between points of interest, ambiguity issues may be reduced. In addition, the embodiment of the invention can record the frequency and time of the attention points of the user, thereby describing the preference degree of the user.
The embodiment of the invention also provides a category determining device of the attention point. Referring to fig. 3, fig. 3 is a schematic structural diagram of a category determining device of an attention point according to an embodiment of the present invention, including:
an obtaining module 310, configured to obtain a predetermined attention point of a single user, and obtain each associated attention point of the predetermined attention point;
a determining module 320, configured to determine, according to a global probability that the predetermined attention point belongs to a predetermined category, a global probability that the respective associated attention point belongs to the predetermined category, and a degree of association between the predetermined attention point and the respective associated attention point, a personalized probability that the predetermined attention point belongs to the predetermined category; wherein the global probability is: in the behavior data of the global user, the probability that the preset attention point or the related attention point belongs to the preset category is determined; the individualization probability is as follows: in the behavior data of the single user, the probability that the predetermined point of interest belongs to the predetermined category.
In one possible implementation, the determining module 320 obtains the global probability from a global point of interest network; the global point of interest network comprises: in the behavior data of the global user, probabilities that all the concerned points belong to different categories are provided.
Fig. 4 is a schematic structural diagram of a category determining device according to an embodiment of the present invention, including: an acquisition module 310, a determination module 320, and a personalized point of interest network determination module 400;
the functions of the obtaining module 310 and the determining module 320 are the same as those of the corresponding modules in the above embodiments, and are not described herein.
A personalized attention point network determining module 400, configured to determine, according to each behavior data of the single user, a directed edge formed by N attention points included in each behavior data; the N is an integer greater than 1; assigning the weight of the directed edge as the number of times that the directed edge appears in each behavior data; according to any two adjacent attention points in each directed edge, determining an undirected edge formed by any two adjacent attention points; assigning the weight of the undirected edge as the sum of the weights of the related directed edges; wherein the related directed edge is a directed edge comprising the arbitrary two adjacent points of interest.
In one possible implementation, the association degree is: and the weight of the undirected edge formed by the preset attention point and the associated attention point.
In a possible implementation manner, the determining module 320 is configured to determine the personalized probability that the predetermined point of interest belongs to the predetermined category by using the following equation:
wherein,
the P' (X, S) is the individualization probability that the preset attention point X belongs to the preset category S;
the P (X, S) is the global probability that the predetermined focus X belongs to the predetermined category S;
said P (Y) i S) is the point of interest Y i Global probabilities belonging to a predetermined category S; the point of interest Y i An associated point of interest for the predetermined point of interest X;
the W is i For the focus X and the focus Y i The weight of the formed undirected edge;
the N is the number of associated attention points of the preset attention point X;
the a and the b are preset parameters.
In one possible implementation, the determining module 320 is further configured to: determining the individuation probability that the preset attention point belongs to other categories; according to the personalized probability that the preset attention point belongs to the preset category and the personalized probability that the preset attention point belongs to other categories, normalized exponential function calculation is carried out on the personalized probability that the preset attention point belongs to the preset category, and the value of the personalized probability that the preset attention point belongs to the preset category is modified to be the result of the calculation.
In a possible implementation manner, the personalized focus network determining module 400 is further configured to determine a timestamp of the directed edge according to a last update time of the behavior data corresponding to the directed edge;
as shown in fig. 4, the apparatus may further include:
a prediction module 430, configured to predict a user behavior of the single user according to the directed edge and the timestamp of the directed edge.
The functions of each module in each device of the embodiments of the present invention may be referred to the corresponding descriptions in the above methods, and are not described herein again.
The embodiment of the invention also provides a category determining device of the attention point, as shown in fig. 5, which is a schematic structural diagram of the category determining device of the attention point of the embodiment of the invention, and includes:
memory 11 and processor 12, memory 11 storing a computer program executable on processor 12. The processor 12, when executing the computer program, implements the category determination method of the point of interest in the above-described embodiment. The number of memories 11 and processors 12 may be one or more.
The apparatus may further include:
and the communication interface 13 is used for communicating with external equipment and carrying out data exchange transmission.
The memory 11 may comprise a high-speed RAM memory or may further comprise a non-volatile memory (non-volatile memory), such as at least one disk memory.
If the memory 11, the processor 12 and the communication interface 13 are implemented independently, the memory 11, the processor 12 and the communication interface 13 may be connected to each other and perform communication with each other through buses. The bus may be an industry standard architecture (ISA, industry Standard Architecture) bus, a peripheral component interconnect (PCI, peripheral Component Interconnect) bus, or an extended industry standard architecture (EISA, extended Industry Standard Architecture) bus, among others. The buses may be classified as address buses, data buses, control buses, etc. For ease of illustration, only one thick line is shown in fig. 5, and not only one bus or one type of bus.
Alternatively, in a specific implementation, if the memory 11, the processor 12 and the communication interface 13 are integrated on a chip, the memory 11, the processor 12 and the communication interface 13 may complete communication with each other through internal interfaces.
An embodiment of the present invention provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements a method as in any of the embodiments included in the first embodiment.
An embodiment of the invention provides a computer program product comprising a computer program which, when executed by a processor, implements a method as in any of the embodiments comprised by.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and further implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
Those of ordinary skill in the art will appreciate that all or a portion of the steps carried out in the method of the above-described embodiments may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, and where the program, when executed, includes one or a combination of the steps of the method embodiments.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing module, or each unit may exist alone physically, or two or more units may be integrated in one module. The integrated modules may be implemented in hardware or in software functional modules. The integrated modules may also be stored in a computer readable storage medium if implemented in the form of software functional modules and sold or used as a stand-alone product. The storage medium may be a read-only memory, a magnetic or optical disk, or the like.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that various changes and substitutions are possible within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (12)
1. A method for determining a category of a point of interest, comprising:
acquiring a preset attention point of a single user, and acquiring each associated attention point of the preset attention point;
determining the individuation probability that the preset attention point belongs to the preset category according to the global probability that the preset attention point belongs to the preset category, the global probability that each associated attention point belongs to the preset category and the association degree of the preset attention point and each associated attention point; wherein,
the global probability is: in the behavior data of the global user, the probability that the preset attention point or the related attention point belongs to the preset category is determined;
the individualization probability is as follows: in the behavior data of the single user, a probability that the predetermined point of interest belongs to the predetermined category;
before determining the personalized probability that the predetermined attention point belongs to the predetermined category, the method further comprises:
according to the behavior data of the single user, determining a directed edge formed by N attention points contained in each behavior data; the N is an integer greater than 1; assigning the weight of the directed edge as the number of times that the directed edge appears in each behavior data; according to any two adjacent attention points in each directed edge, determining an undirected edge formed by any two adjacent attention points; assigning the weight of the undirected edge as the sum of the weights of the related directed edges; wherein the related directed edge is a directed edge comprising the arbitrary two adjacent points of interest;
the association degree is as follows: and the weight of the undirected edge formed by the preset attention point and the associated attention point.
2. The method of claim 1, wherein the step of determining the position of the substrate comprises,
the global probability is obtained from a global attention point network;
the global point of interest network comprises: in the behavior data of the global user, probabilities that all the concerned points belong to different categories are provided.
3. The method of claim 1, wherein the determining the personalized probability that the predetermined point of interest belongs to the predetermined category based on the global probability that the predetermined point of interest belongs to the predetermined category, the global probability that the respective associated point of interest belongs to the predetermined category, and the degree of association of the predetermined point of interest with the respective associated point of interest comprises:
the following equation is used for determination:
wherein, the method comprises the steps of, wherein,
the saidThe personalized probability that the predetermined focus X belongs to the predetermined category S;
the saidGlobal probability that a predetermined focus X belongs to a predetermined category S;
the saidFor the focus->Global probabilities belonging to a predetermined category S; the focus->An associated point of interest for the predetermined point of interest X;
the saidFor the point of interest X and the point of interest->The weight of the formed undirected edge;
the N is the number of associated attention points of the preset attention point X;
the a and the b are preset parameters.
4. A method according to any one of claims 1 to 3, further comprising:
determining the individuation probability that the preset attention point belongs to other categories;
according to the personalized probability that the preset attention point belongs to the preset category and the personalized probability that the preset attention point belongs to other categories, normalized exponential function calculation is carried out on the personalized probability that the preset attention point belongs to the preset category, and the value of the personalized probability that the preset attention point belongs to the preset category is modified to be the result of the calculation.
5. The method of claim 1, wherein after determining the directed edges of the N points of interest included in each behavior data, further comprising: determining a time stamp of the directed edge according to the last update time of the behavior data corresponding to the directed edge;
the method further comprises the steps of: and predicting the user behavior of the single user according to the directed edge and the time stamp of the directed edge.
6. A category determining apparatus of a point of interest, comprising:
the acquisition module is used for acquiring the preset attention point of a single user and acquiring each associated attention point of the preset attention point;
the determining module is used for determining the individuation probability that the preset attention point belongs to the preset category according to the global probability that the preset attention point belongs to the preset category, the global probability that each associated attention point belongs to the preset category and the association degree of the preset attention point and each associated attention point; wherein the global probability is: in the behavior data of the global user, the probability that the preset attention point or the related attention point belongs to the preset category is determined; the individualization probability is as follows: in the behavior data of the single user, a probability that the predetermined point of interest belongs to the predetermined category;
the apparatus further comprises:
the personalized attention point network determining module is used for determining a directed edge formed by N attention points contained in each behavior data according to the behavior data of the single user; the N is an integer greater than 1; assigning the weight of the directed edge as the number of times that the directed edge appears in each behavior data; according to any two adjacent attention points in each directed edge, determining an undirected edge formed by any two adjacent attention points; assigning the weight of the undirected edge as the sum of the weights of the related directed edges; wherein the related directed edge is a directed edge comprising the arbitrary two adjacent points of interest;
the association degree is as follows: and the weight of the undirected edge formed by the preset attention point and the associated attention point.
7. The apparatus of claim 6, wherein the determination module obtains the global probability from a global point of interest network;
the global point of interest network comprises: in the behavior data of the global user, probabilities that all the concerned points belong to different categories are provided.
8. The apparatus of claim 6, wherein the means for determining determines the personalized probability that the predetermined point of interest belongs to the predetermined category using the following equation:
wherein, the method comprises the steps of, wherein,
the saidThe personalized probability that the predetermined focus X belongs to the predetermined category S;
the saidGlobal probability that a predetermined focus X belongs to a predetermined category S;
the saidFor the focus->Global probabilities belonging to a predetermined category S; the focus->An associated point of interest for the predetermined point of interest X;
the saidFor the point of interest X and the point of interest->The weight of the formed undirected edge;
the N is the number of associated attention points of the preset attention point X;
the a and the b are preset parameters.
9. The apparatus of any one of claims 6 to 8, wherein the determining module is further configured to: determining the individuation probability that the preset attention point belongs to other categories; according to the personalized probability that the preset attention point belongs to the preset category and the personalized probability that the preset attention point belongs to other categories, normalized exponential function calculation is carried out on the personalized probability that the preset attention point belongs to the preset category, and the value of the personalized probability that the preset attention point belongs to the preset category is modified to be the result of the calculation.
10. The apparatus of claim 6, wherein the personalized focus network determination module is further configured to determine a timestamp of the directed edge based on a last update time of the behavior data corresponding to the directed edge;
the apparatus further comprises: and the prediction module is used for predicting the user behavior of the single user according to the directed edge and the time stamp of the directed edge.
11. A category determining device of a point of interest, the device comprising:
one or more processors;
a storage means for storing one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 1-5.
12. A computer readable storage medium storing a computer program, which when executed by a processor implements the method of any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910572735.4A CN112148960B (en) | 2019-06-27 | 2019-06-27 | Method, device, equipment and storage medium for determining category of attention point |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910572735.4A CN112148960B (en) | 2019-06-27 | 2019-06-27 | Method, device, equipment and storage medium for determining category of attention point |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112148960A CN112148960A (en) | 2020-12-29 |
| CN112148960B true CN112148960B (en) | 2024-03-22 |
Family
ID=73869163
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910572735.4A Active CN112148960B (en) | 2019-06-27 | 2019-06-27 | Method, device, equipment and storage medium for determining category of attention point |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112148960B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2312515A1 (en) * | 2009-10-16 | 2011-04-20 | Alcatel Lucent | Device for determining potential future interests to be introduced into profile(s) of user(s) of communication equipment(s) |
| CN102609528A (en) * | 2012-02-14 | 2012-07-25 | 云南大学 | Frequent mode association sorting method based on probabilistic graphical model |
| WO2013157705A1 (en) * | 2012-04-17 | 2013-10-24 | 한국과학기술원 | Method for inferring interest of user through interests of social neighbors and topics of social activities in sns, and system therefor |
| CN103399883A (en) * | 2013-07-19 | 2013-11-20 | 百度在线网络技术（北京）有限公司 | Method and system for performing personalized recommendation according to user interest points/concerns |
| CN107301248A (en) * | 2017-07-19 | 2017-10-27 | 百度在线网络技术（北京）有限公司 | Term vector construction method and device, computer equipment, the storage medium of text |
| CN108647732A (en) * | 2018-05-14 | 2018-10-12 | 北京邮电大学 | A kind of pathological image sorting technique and device based on deep neural network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8392343B2 (en) * | 2010-07-21 | 2013-03-05 | Yahoo! Inc. | Estimating probabilities of events in sponsored search using adaptive models |
-
2019
- 2019-06-27 CN CN201910572735.4A patent/CN112148960B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2312515A1 (en) * | 2009-10-16 | 2011-04-20 | Alcatel Lucent | Device for determining potential future interests to be introduced into profile(s) of user(s) of communication equipment(s) |
| CN102609528A (en) * | 2012-02-14 | 2012-07-25 | 云南大学 | Frequent mode association sorting method based on probabilistic graphical model |
| WO2013157705A1 (en) * | 2012-04-17 | 2013-10-24 | 한국과학기술원 | Method for inferring interest of user through interests of social neighbors and topics of social activities in sns, and system therefor |
| CN103399883A (en) * | 2013-07-19 | 2013-11-20 | 百度在线网络技术（北京）有限公司 | Method and system for performing personalized recommendation according to user interest points/concerns |
| CN107301248A (en) * | 2017-07-19 | 2017-10-27 | 百度在线网络技术（北京）有限公司 | Term vector construction method and device, computer equipment, the storage medium of text |
| CN108647732A (en) * | 2018-05-14 | 2018-10-12 | 北京邮电大学 | A kind of pathological image sorting technique and device based on deep neural network |
Non-Patent Citations (2)
| Title |
|---|
| SMART:一种面向电商平台快速消费品的图推荐算法;卿勇;刘梦娟;银盈;李杨曦;;计算机科学;20171115(S2);全文 * |
| 关注点推荐算法的霍克斯过程法;张国明;王俊淑;江南;盛业华;;测绘学报;20180915(09);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112148960A (en) | 2020-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TW202008237A (en) | Method and device for training prediction model for new scenario | |
| CN108363792A (en) | Link generation method, device, storage medium and the electronic equipment of five application page | |
| CN105549936B (en) | Data table display method and device | |
| CN104281455A (en) | Interface calling method and device and terminal | |
| CN110069739B (en) | Page preloading method and device | |
| CN107958039A (en) | A kind of term error correction method, device and server | |
| CN109478311A (en) | A kind of image-recognizing method and terminal | |
| CN110020383B (en) | Page data request processing method and device | |
| CN109710396A (en) | A kind of method and device of information collection and memory release | |
| CN107247722A (en) | File scanning method and device and intelligent terminal | |
| CN110298508A (en) | Behavior prediction method, device and equipment | |
| CN112395322A (en) | List data display method and device based on hierarchical cache and terminal equipment | |
| CN107885875B (en) | Synonymy transformation method and device for search words and server | |
| CN109271564A (en) | Declaration form querying method and equipment | |
| CN113205189B (en) | Method for training prediction model, prediction method and device | |
| CN112148960B (en) | Method, device, equipment and storage medium for determining category of attention point | |
| CN104166649A (en) | Caching method and device for search engine | |
| CN117131540A (en) | Task processing method and device based on large language model and electronic equipment | |
| WO2014117594A1 (en) | Method and device for updating web page | |
| CN116226666A (en) | Rule conflict detection method and device, electronic equipment and readable storage medium | |
| US20220284060A1 (en) | Question Answering Method and Apparatus Based on Knowledge Graph | |
| CN111666020B (en) | Method and device for switching document pages, electronic equipment and storage medium | |
| CN110223108B (en) | Click through rate prediction method, device and equipment | |
| CN110059272B (en) | Page feature recognition method and device | |
| CN112256820B (en) | Document positioning method and device |
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 |