WO2016134599A1 - 一种用于确定标杆对象的方法和装置 - Google Patents

一种用于确定标杆对象的方法和装置 Download PDF

Info

Publication number
WO2016134599A1
WO2016134599A1 PCT/CN2015/092881 CN2015092881W WO2016134599A1 WO 2016134599 A1 WO2016134599 A1 WO 2016134599A1 CN 2015092881 W CN2015092881 W CN 2015092881W WO 2016134599 A1 WO2016134599 A1 WO 2016134599A1
Authority
WO
WIPO (PCT)
Prior art keywords
objects
transfer
information
keywords
determining
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.)
Ceased
Application number
PCT/CN2015/092881
Other languages
English (en)
French (fr)
Inventor
周泽伟
程涛远
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.)
Baidu Online Network Technology Beijing Co Ltd
Original Assignee
Baidu Online Network Technology Beijing 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 Baidu Online Network Technology Beijing Co Ltd filed Critical Baidu Online Network Technology Beijing Co Ltd
Publication of WO2016134599A1 publication Critical patent/WO2016134599A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor

Definitions

  • the present invention relates to the field of computer technologies, and in particular, to a method and apparatus for determining a target object.
  • the computer can only assist in manually determining the target object of the object by sorting the data from the object perspective, such as the industry and region to which the object belongs, the sales situation of the object, and the market demand.
  • a method for determining a benchmark object for an object in a computer device comprises:
  • transfer situation information of the plurality of objects where the transfer situation information is used to indicate a transfer situation of the user in the plurality of objects based on the object information acquisition behavior;
  • an apparatus for determining a benchmark object for an object in a computer device comprising:
  • the benchmark object determined by the solution of the present invention is more objective and accurate; 4) even in the data from the user's point of view, the transfer situation information of the present invention is not common data, in fact, if explicitly And data from the user's point of view, those skilled in the art are more likely to think of direct evaluations from users (such as scoring, commenting, etc.).
  • FIG. 1 is a schematic flow chart of a method for determining a target object according to a preferred embodiment of the present invention
  • FIG. 2 is a schematic structural diagram of a determining apparatus for determining a target object according to a preferred embodiment of the present invention
  • Figure 3 is a diagram showing a transfer path of a user in a plurality of objects in a preferred embodiment
  • FIG. 5 shows a specific example of transition from a transfer path of a mesh structure of a keyword to a transfer path of a mesh structure of an object
  • Figure 6 shows a schematic diagram of a transfer from one object to a plurality of objects in a preferred embodiment
  • the method of this embodiment is mainly implemented by a computer device; the computer device includes a network device and a user device.
  • the network device includes, but is not limited to, a single network server, a server group composed of multiple network servers, or a cloud computing-based cloud composed of a large number of computers or network servers, where the cloud computing is distributed computing.
  • a super virtual computer consisting of a group of loosely coupled computers; the network in which the network device is located includes, but is not limited to, the Internet, a wide area network, a metropolitan area network, a local area network, a VPN network, and the like.
  • the user equipment includes, but is not limited to, a PC, a tablet, a smart phone, a PDA, an IPTV, and the like.
  • the method according to the present embodiment includes step S1 and step S2.
  • step S1 the computer device acquires transfer status information of a plurality of objects.
  • the object is of a commercial nature. More preferably, the object comprises a brand.
  • the transfer situation information is used to indicate a transfer situation of the user in the plurality of objects based on the object information acquisition behavior.
  • the object information acquisition behavior includes any behavior that can be used to acquire information of the object; for example, the object information acquisition behavior includes an action of acquiring object information by searching for a keyword related to the object; for example, the object Information acquisition behavior includes the behavior of obtaining object information by clicking and browsing content related to the object.
  • the “object-based information acquisition behavior” indicates that the transfer situation reflects a transfer situation generated by the user in the object information acquisition behavior.
  • the transfer situation needs to be determined based on the object information acquisition behavior; for example, The user is determined to be in the middle object by counting the search objects that are changed by the user in the search behavior, or by counting the search keywords associated with the objects that are changed by the plurality of users in the search behavior. Transfer information, etc.
  • the transfer situation information of the plurality of objects includes but is not limited to at least one of the following:
  • the transfer path information indicates a transfer path of the user among the plurality of objects. For example, there are three objects Object1, Object2, and Object3, and the transfer path information indicates that the transfer paths of a plurality of users among the three objects include: transfer from Object1 to Object2, and transfer from Object1 to Object3.
  • the transfer number information indicates the number of times the user transfers between the respective objects. For example, there are three objects Object1, Object2, and Object3, and the number of transitions indicates that the number of transitions of a plurality of users between the three objects includes: five times from Object1 to Object2, and eight times from Object1 to Object3.
  • the transition probability information indicates a transition probability of the user between the respective objects. For example, there are three objects Object1, Object2, and Object3, and the transition probability information indicates that the transition probabilities of multiple users between the three objects include: the probability of transferring from Object1 to Object2 is 38.46%, and the transfer from Object1 to Object3 The probability is 61.54%.
  • the transfer situation information may adopt multiple storage modes.
  • the transfer situation information is stored as a table, and the transfer path of the user in the plurality of objects and the number of transitions and transition probabilities of the user between the respective objects are recorded in the table, as shown in Table 1 below.
  • the transfer situation information includes: a transfer path stored as a mesh structure, and a transfer number and/or a transfer probability between respective nodes (ie, between respective objects) in the mesh structure.
  • the transfer situation information of the 9 objects includes the transfer path as shown in FIG. 3, and between the nodes having the arrow connection in FIG. 3 (eg, from Object1 to Object2, from Object1 to The number of transitions and/or transition probabilities for Object3, from Object1 to Object4, and so on.
  • the manner in which the computer device acquires the transfer situation information of the multiple objects includes but is not limited to:
  • the computer device directly obtains the pre-determined transfer situation information of the plurality of objects.
  • the computer device reads the pre-determined transfer situation information for the plurality of objects from a local or other device.
  • Step S1 further includes step S11 and step S12.
  • step S11 the computer device acquires transfer situation information of a plurality of keywords.
  • the transfer situation information of the plurality of keywords is used to indicate a transfer situation of the user in the plurality of keywords based on the object information acquisition behavior.
  • the plurality of keywords are associated with an object in the object information acquisition behavior; for example, if the object information acquisition behavior is an object search behavior, the keyword may be a search keyword input or selected by the user in the search behavior. Wait.
  • the transfer situation information of the plurality of keywords includes at least one of the following:
  • the transfer path information of the user in the plurality of keywords indicates a transfer path of the user in the plurality of keywords.
  • the transfer path information indicates that the transfer paths of the plurality of users in the three keywords include: transfer from Query1 to Query2, and transfer from Query1 to Query3.
  • the information on the number of times the user transfers between the keywords indicates the number of times the user transfers between the keywords. For example, there are three keywords Query1, Query2, and Query3, and the number of transitions indicates that the number of transitions of the plurality of users between the three keywords includes: five times from Query1 to Query2, and eight times from Query1 to Query3 .
  • some of the plurality of keywords may not have a transfer path between them (that is, the user does not transfer between the partial keywords in the object information acquisition behavior), and the partial keywords are The number of transitions between is zero.
  • the transfer situation information of the plurality of keywords may adopt multiple storage manners.
  • the transfer situation information is stored as a table, and the transfer path of the user among the plurality of keywords and the number of times the user transfers between the respective keywords are recorded in the table, as shown in Table 2 below.
  • the transfer situation information includes: a transfer path stored as a mesh structure, and a number of transfers between the nodes in the mesh structure (ie, between respective keywords).
  • the transfer situation information of the nine objects includes the transfer path as shown in FIG. 4, and the number of transitions between the respective nodes having the arrow connection in FIG.
  • the implementation manner of the computer device acquiring the transfer situation information of multiple keywords but not limited to:
  • the computer device directly obtains the pre-determined transfer situation information of the plurality of keywords.
  • the computer device reads the pre-determined transfer situation information for the plurality of keywords from a local or other device.
  • the computer device acquires a keyword attention record of at least one user, and determines transfer information of the plurality of keywords according to the keyword attention record.
  • the keyword attention record includes a keyword that the plurality of users have paid attention to in the object information acquisition behavior and time information that the keyword is focused on.
  • the object information acquisition behavior includes a search behavior
  • the focused keyword includes a searched keyword;
  • the object information acquisition behavior includes a browsing behavior, and the focused keyword includes being clicked for browsing The keyword of the object content.
  • the computer device determines the user's transfer path in the keyword and the transfer between the keywords according to the time information of the keyword included in the keyword attention record. The number of times; and the computer device determines the transfer situation information of the plurality of keywords by combining the transfer paths of the respective users in the keywords and the number of transitions between the respective keywords.
  • the computer device obtains the keyword attention record of the user A and the user B; wherein the keyword attention records of the user A and the user B are respectively as shown in Tables 3 and 4 below:
  • the computer device determines that user A is in the keyword
  • the transfer path in the middle includes "Query1 ⁇ Query3", and the number of transitions of "Query1 ⁇ Query3" is 1; similarly, the computer device determines that the transfer path of the user B in the keyword includes "Query1 ⁇ Query2", and "Query1 ⁇ Query2" The number of transfers is 1.
  • the computer device merges the transfer paths of the users A and B in the keywords and the number of transitions between the keywords, and determines the transfer situation information of the plurality of keywords as shown in Table 5 below.
  • step S12 the computer device determines the transition situation information of the plurality of objects based on the object to which the plurality of keywords are associated and the transition situation information of the plurality of keywords.
  • the computer device may determine, according to the user's transfer path information in the plurality of keywords and the object to which the plurality of keywords are respectively associated, the transfer path information of the user in the plurality of objects, and the computer device according to the user in each key
  • the number of times of transfer between words and the object to which the plurality of keywords are respectively associated with each other determines the number of times of transfer and/or transition probability information between the users.
  • the transition situation information of the plurality of keywords is as shown in Table 2 above, and Query1, Query2, and Query3 are respectively associated with Object1, Object2, and Object3; and the transfer information of the keyword according to the keyword shown in Table 2 and the foregoing association relationship are performed by the computer device.
  • the computer device can calculate the transition probability p ij of the object i to the object j based on the following formula:
  • a ij represents the number of transitions from object i to object j. Represents the number of transitions from object i to all objects.
  • the object Object0 can be transferred to itself and other objects Object1 to Object13; for the transition from the object Object1 to Object8, the transition probability of the object Objectp0 to the object Object8 among them, Represents the number of transfers from object Object0 to Object0 itself and from Object1 to Object13.
  • FIG. 5 illustrates an example of transition from a transition path of the mesh structure of the keyword to a transition path of the mesh structure of the object.
  • each node in the upper mesh structure is a keyword
  • each node in the lower mesh structure is an object corresponding to a keyword in the corresponding node above.
  • step S2 the computer device determines its benchmark object for at least one of the plurality of objects based on the transition situation information of the plurality of objects obtained in step S1, wherein the benchmark object belongs to the plurality of objects.
  • the plurality of objects belong to the same class do not.
  • an object's benchmark object indicates an object that is competitive with the object, for example, an object that is commercially competitive with the object.
  • the implementation manner of determining, by the computer device, the benchmark object for the at least one of the multiple objects according to the transfer situation information of the multiple objects includes but is not limited to:
  • the computer device determines its benchmark object for at least one of the plurality of objects directly based on the transfer situation information of the plurality of objects.
  • the computer device determines Object3 as the benchmark object of Object1 according to the transition probability from Object1 to Object3 exceeding a predetermined threshold of 60%.
  • the computer device calculates an average of transition probabilities of the other of the plurality of objects to the object; and then, the computer device selects one of the objects with the highest average as the plurality A common benchmark object for other objects in the object.
  • Step S2 further includes step S21 and step S22.
  • step S21 the computer device determines the ordering of the plurality of objects based on the transfer situation information of the plurality of objects.
  • the ordering of the plurality of objects may include a sorting position of each of the plurality of objects in the plurality of objects, or may only include a sorting position of each of the plurality of objects.
  • the implementation manner of determining, by the computer device, the ordering of the multiple objects according to the transfer situation information of the multiple objects includes but is not limited to:
  • the computer device directly determines the plurality of objects according to the transfer situation information of the plurality of objects Sort of.
  • the computer device calculates an average of the transition probabilities of other of the plurality of objects to the object; then, the computer device averages the transition probabilities for each object, Sort multiple objects.
  • the computer device determines the ordering of the plurality of objects based on the transfer situation information of the plurality of objects and the initial ordering of the plurality of objects.
  • the initial ordering may be manually set or pre-determined by the transfer situation information of multiple objects.
  • the computer device may adjust an initial order of the multiple objects according to the transfer situation information of the multiple objects to obtain an order of the multiple objects.
  • This implementation includes but is not limited to:
  • the computer device directly adjusts the initial ordering of the plurality of objects according to the transfer situation information of the plurality of objects, and obtains the sorting of the plurality of objects.
  • the computer device shifts the probability of shifting to another object according to the number of transitions and/or the transition probability included in the transition situation information, and the order of the objects with higher probability of transferring other objects to themselves is up-regulated.
  • the computer device repeatedly performs the following steps until the predetermined condition is met: the computer device determines the current ordering of the plurality of objects according to the transfer situation information of the plurality of objects and the sorting of the plurality of objects determined last time; Preferably, the computer device may determine the current ordering of the plurality of objects based on a matrix of the transfer situation information of the object and a matrix of the plurality of objects determined last time, such as by multiplying the two matrices. Wherein the last determined order is the ordering of the plurality of objects; wherein, when the step of determining the current ordering of the plurality of objects is performed for the first time, the plurality of objects of the last determined Sorted to the initial sort.
  • the predetermined condition includes any condition capable of terminating the repeating step.
  • the predetermined condition includes but is not limited to at least one of the following:
  • the number of repetitions reaches a predetermined repetition threshold; for example, the predetermined repetition threshold is 50 times, and the plurality of objects are determined when the computer device performs 50 times of the transition situation information according to the plurality of objects and the ranking of the plurality of objects determined last time. After this sorting operation, stop the repeat operation.
  • the ordering of the plurality of objects does not change; for example, when the computer device repeatedly performs the transfer information according to the plurality of objects 5 times And after the ranking of the plurality of objects determined last time to determine the current ordering of the plurality of objects, the obtained ordering of the plurality of objects does not change, and the computer device stops the repeating operation.
  • the computing device can calculate the current ordering of the plurality of objects according to the following formula:
  • R t is the current ordering of the plurality of objects
  • R t-1 is the ordering of the plurality of objects determined last time
  • l(p i , p j ) is the transition probability of object i to object j, where 1 ⁇ i ⁇ N, 1 ⁇ j ⁇ N, l (p i , p j ) can be calculated according to the transfer situation information, as calculated according to the following formula:
  • b ij represents the number of transitions from object i to object j
  • the sum of the probabilities of transferring from one object to all of the plurality of objects is 1, that is, l(p i , p j ) satisfies the following formula:
  • the initial ordered state matrix R 0 of the plurality of objects is as follows:
  • PageRank(p i ) 0 represents an initial sorting value of the object i
  • the PageRank(p i ) 0 can be manually set or can be determined according to the transition situation information.
  • the PageRank(p i ) 0 can be determined according to the following formula:
  • the above examples are only for better explaining the technical solutions of the present invention, and are not intended to limit the present invention.
  • the ordering of the objects, determining the implementation of the current ordering of the plurality of objects should be included in the scope of the present invention; for example, in the above formula for calculating R t , the offset matrix may adopt other values. , such as other manually set values, etc.; for example, if all l(p i , p j ) are not zero, then the above formula for calculating R t may not require the offset matrix, ie, the formula may tweak to:
  • step S22 the computer device determines its benchmark object for at least one of the plurality of objects according to the ordering of the plurality of objects.
  • a computer device treats the top Z objects of the plurality of objects as the common benchmark object of the objects after Z. Where Z is a preset natural number. When Z is 1, the computer device uses the highest ranked object as a common benchmark object for all other objects in the plurality of objects.
  • the computer device takes an object whose sorting position is close to the sorting position of the object as a benchmark object of the object or the like.
  • the computer can only assist in manually determining the target object of the object by sorting the data from the object perspective, such as the industry and region to which the object belongs, the sales situation of the object, and the market demand.
  • the present invention provides a solution for a computer device to automatically determine a benchmark object; and, the present invention breaks the bias of those skilled in the art and can determine a benchmark object by analyzing the user's transitions among multiple objects.
  • the scheme of determining the benchmark object is closer to the user angle, and the user's understanding of the benchmark object is more intuitively reflected. Therefore, the benchmark object determined by the solution of the present invention is more objective and accurate; moreover, even in the data from the user's point of view, the transfer situation information of the present invention is not common data. In fact, if the data from the user's point of view is explicitly mentioned, those skilled in the art are more likely to think of direct evaluations from users (such as scoring, commenting, etc.).
  • the determining device may be installed in a computer device, the determining device including means for acquiring transfer situation information of a plurality of objects (hereinafter referred to as "acquisition device 1"), and configured to use the plurality of objects according to the transfer situation information At least one of the objects determines a device of its target object (hereinafter referred to as "sub-determination device 2").
  • the acquisition device 1 acquires transfer situation information of a plurality of objects.
  • the object is of a commercial nature. More preferably, the object comprises a brand.
  • the transfer situation information is used to indicate a transfer situation of the user in the plurality of objects based on the object information acquisition behavior.
  • the object information acquisition behavior includes any behavior that can be used to acquire information of the object; for example, the object information acquisition behavior includes an action of acquiring object information by searching for a keyword related to the object; for example, the object Information acquisition behavior includes the behavior of obtaining object information by clicking and browsing content related to the object.
  • the “object-based information acquisition behavior” indicates that the transfer situation reflects a transfer situation generated by the user in the object information acquisition behavior.
  • the transfer situation needs to be determined based on the object information acquisition behavior; for example, The information about the transfer of the user in the middle object is determined by counting the search objects that are changed by the plurality of users in the search behavior, or by counting the search keywords associated with the objects that are changed by the plurality of users in the search behavior.
  • the transfer situation information of the plurality of objects includes but is not limited to at least one of the following:
  • the transfer path information indicates a transfer path of the user among the plurality of objects. For example, there are three objects Object1, Object2, and Object3, and the transfer path information indicates that the transfer paths of a plurality of users among the three objects include: transfer from Object1 to Object2, and transfer from Object1 to Object3.
  • the transfer number information indicates the number of times the user transfers between the respective objects. For example, there are three objects Object1, Object2, and Object3, and the number of transitions indicates that the number of transitions of a plurality of users between the three objects includes: five times from Object1 to Object2, and eight times from Object1 to Object3.
  • the transition probability information indicates a transition probability of the user between the respective objects. For example, there are three objects Object1, Object2, and Object3, and the transition probability information indicates that the transition probabilities of multiple users between the three objects include: the probability of transferring from Object1 to Object2 is 38.46%, and the transfer from Object1 to Object3 The probability is 61.54%.
  • the transfer situation information may adopt multiple storage modes.
  • the transfer situation information is stored as a table, and the transfer path of the user in the plurality of objects and the number of transitions and transition probabilities of the user between the respective objects are recorded in the table, as shown in Table 1 above.
  • the transfer situation information includes: a transfer path stored as a mesh structure, and a transfer number and/or a transfer probability between respective nodes (ie, between respective objects) in the mesh structure.
  • the transfer situation information of the 9 objects includes the transfer path as shown in FIG. 3, and between the nodes having the arrow connection in FIG. 3 (eg, from Object1 to Object2, from Object1 to The number of transitions and/or transition probabilities for Object3, from Object1 to Object4, and so on.
  • the manner in which the acquiring device 1 acquires the transfer situation information of the multiple objects includes but is not limited to:
  • the acquiring device 1 directly acquires the pre-determined transfer situation information of the plurality of objects.
  • the acquisition device 1 reads the pre-determined transfer situation information of the plurality of objects from a local or other device.
  • the acquisition device 1 further includes means for acquiring transfer situation information of a plurality of keywords (hereinafter referred to as “sub-acquisition device”, not shown) and objects for respectively associated with the plurality of keywords and the plurality of A device for transferring the situation of the keywords, and means for determining the transfer situation information of the plurality of objects (hereinafter referred to as "transfer situation determining means”, not shown).
  • the sub-acquisition device acquires transfer situation information of a plurality of keywords.
  • the transfer situation information of the plurality of keywords is used to indicate a transfer situation of the user in the plurality of keywords based on the object information acquisition behavior.
  • the plurality of keywords are associated with an object in the object information acquisition behavior; for example, if the object information acquisition behavior is an object search behavior, the keyword may be a search keyword input or selected by the user in the search behavior. Wait.
  • the transfer situation information of the plurality of keywords includes at least one of the following:
  • the transfer path information of the user in the plurality of keywords indicates a transfer path of the user in the plurality of keywords.
  • the transfer path information indicates that the transfer paths of the plurality of users among the three keywords include: transition from Query1 to Query2, and transfer from Query1 to Query3.
  • the information on the number of times the user transfers between the keywords indicates the number of times the user transfers between the keywords. For example, there are three keywords Query1, Query2, and Query3, and the number of transitions indicates that the number of transitions of the plurality of users between the three keywords includes: five times from Query1 to Query2, and eight times from Query1 to Query3 .
  • the transfer situation information of the plurality of keywords may adopt multiple storage manners.
  • the transfer situation information is stored as a table, and the transfer path of the user among the plurality of keywords and the number of times the user transfers between the respective keywords are recorded in the table, as shown in Table 2 above.
  • the transfer situation information includes: a transfer path stored as a mesh structure, and a number of transfers between the nodes in the mesh structure (ie, between respective keywords).
  • the transfer situation information of the nine objects includes the transfer path as shown in FIG. 4, and the number of transitions between the respective nodes having the arrow connection in FIG.
  • the implementation manner in which the sub-acquisition device acquires the transfer situation information of the multiple keywords includes but is not limited to:
  • the sub-acquisition device directly acquires the pre-determined transfer situation information of the plurality of keywords.
  • the sub-acquisition device reads the pre-determined transfer situation information of the plurality of keywords from a local or other device.
  • the sub-acquisition device acquires a keyword attention record of at least one user, and determines transfer situation information of the plurality of keywords according to the keyword attention record.
  • the keyword attention record includes a keyword that the plurality of users have paid attention to in the object information acquisition behavior and time information that the keyword is focused on.
  • the object information acquisition behavior includes a search behavior
  • the focused keyword includes a searched keyword;
  • the object information acquisition behavior includes a browsing behavior, and the focused keyword includes being clicked for browsing The keyword of the object content.
  • the sub-acquisition device determines the user's keyword in accordance with the time information of the keyword included in the keyword attention record.
  • the sub-acquisition device acquires the keyword attention record of the user A and the user B; wherein the keyword attention records of the user A and the user B are respectively shown in Tables 3 and 4 above.
  • the sub-acquisition device determines that the transfer path of the user A in the keyword includes “Query1 ⁇ Query3”, and the number of transitions of “Query1 ⁇ Query3” is 1; similarly, the sub-acquisition device determines the user The transfer path of B in the keyword includes "Query1 ⁇ Query2", and the number of transitions of "Query1 ⁇ Query2" is 1.
  • the sub-acquisition device merges the transfer paths of the users A and B in the keywords and the number of transitions between the keywords, and determines the transfer situation information of the plurality of keywords as shown in Table 5 above.
  • the transition situation determining means determines the transition situation information of the plurality of objects based on the objects to which the plurality of keywords are associated and the transition situation information of the plurality of keywords.
  • the transition situation determining means may determine the transfer path information of the user in the plurality of objects according to the transfer path information of the user in the plurality of keywords and the object to which the plurality of keywords are respectively associated, and the transfer situation determining means is configured according to The information on the number of times of transfer between the keywords and the object to which the plurality of keywords are respectively associated with each other determines the number of times of transfer and/or transition probability information between the users.
  • the transition situation determining means may calculate the transition probability p ij of the object i to the object j based on the following formula:
  • a ij represents the number of transitions from object i to object j. Represents the number of transitions from object i to all objects.
  • the object Object0 can be transferred to itself and other objects Object1 to Object13; for the transition from the object Object1 to Object8, the transition probability of the object Objectp0 to the object Object8 among them, Represents the number of transfers from object Object0 to Object0 itself and from Object1 to Object13.
  • FIG. 5 illustrates an example of transition from a transition path of the mesh structure of the keyword to a transition path of the mesh structure of the object.
  • each node in the upper mesh structure is a keyword
  • each node in the lower mesh structure is an object corresponding to a keyword in the corresponding node above.
  • the sub-determination device 2 determines a benchmark object for at least one of the plurality of objects based on the transfer situation information of the plurality of objects obtained by the acquisition device 1, wherein the benchmark object For the plurality of objects.
  • the plurality of objects belong to the same category.
  • an object's benchmark object indicates an object that is competitive with the object, for example, an object that is commercially competitive with the object.
  • the sub-determining device 2 determines, according to the transition situation information of the plurality of objects, an implementation manner of determining the benchmark object for the at least one of the plurality of objects, including but not limited to:
  • the sub-determination device 2 determines its benchmark object for at least one of the plurality of objects directly based on the transfer situation information of the plurality of objects.
  • the sub-determination device 2 determines Object3 as the target object of Object1 based on the transition probability from Object1 to Object3 exceeding a predetermined threshold of 60%.
  • the sub-determination device 2 calculates an average value of transition probabilities of the other of the plurality of objects to the object; and then, the sub-determining device 2 selects an object having the highest average value, A common benchmark object that is the other of the plurality of objects.
  • the sub-determining device 2 further includes means for determining the ordering of the plurality of objects based on the transfer situation information (hereinafter referred to as “sorting determining means", not shown), for A device for determining at least one of the plurality of objects to determine its target object (hereinafter referred to as “the benchmark object determining device”, not shown).
  • the sort determining means determines the ordering of the plurality of objects based on the transfer situation information of the plurality of objects.
  • the ordering of the plurality of objects may include a sorting position of each of the plurality of objects in the plurality of objects, or may only include a sorting position of each of the plurality of objects.
  • the order determining device determines, according to the transfer situation information of the plurality of objects, an implementation manner of the sorting of the multiple objects, including but not limited to:
  • the sort determining means determines the ordering of the plurality of objects directly based on the transfer situation information of the plurality of objects.
  • the ranking determining means calculates an average of the transition probabilities of the other of the plurality of objects to the object; and then, the ranking determining means averages the transition probabilities according to each of the objects Value, sorting multiple objects.
  • the sort determining means determines the ordering of the plurality of objects based on the transfer situation information of the plurality of objects and the initial ordering of the plurality of objects.
  • the initial ordering may be manually set or pre-determined by the transfer situation information of multiple objects.
  • the order determining device may adjust the initial ordering of the plurality of objects according to the transfer situation information of the plurality of objects to obtain the sorting of the plurality of objects.
  • This implementation includes but is not limited to:
  • the sort determining means directly adjusts the initial sorting of the plurality of objects according to the transfer situation information of the plurality of objects, and obtains the sorting of the plurality of objects.
  • the ranking determining means shifts the ranking to the other objects based on the number of transitions and/or the transition probability included in the transition situation information, and the ranking of the objects with higher probability of transition of other objects to themselves is up-regulated.
  • the sort determining means repeatedly performs the following operations until the predetermined condition is satisfied: the sort determining means determines the current sorting of the plurality of objects based on the transfer situation information of the plurality of objects and the sorting of the plurality of objects determined last time More preferably, the order determining means may determine the current ordering of the plurality of objects according to the matrix of the transfer condition information of the object and the ordered matrix of the plurality of objects determined last time, such as by multiplying the two matrices . Wherein the last determined ranking is the ordering of the plurality of objects; wherein, when it is determined that the current sorting operation of the plurality of objects is performed for the first time, the plurality of the last determined The ordering of the objects is the initial ordering.
  • the predetermined condition includes any condition capable of terminating the repeated operation.
  • the predetermined condition includes but is not limited to at least one of the following:
  • the number of repetitions reaches a predetermined repetition threshold; for example, the predetermined repetition threshold is 50 times, and the ranking determining means performs 50 times to determine the plurality of pieces based on the transition situation information of the plurality of objects and the sorting of the plurality of objects determined last time After the current sorting operation of the object, the repeat operation is stopped.
  • the predetermined repetition threshold is 50 times
  • the ranking determining means performs 50 times to determine the plurality of pieces based on the transition situation information of the plurality of objects and the sorting of the plurality of objects determined last time After the current sorting operation of the object, the repeat operation is stopped.
  • the ordering of the plurality of objects does not change; for example, when the sort determining means repeatedly performs the transfer according to the plurality of objects 5 times After the information and the sorting of the plurality of objects determined last time to determine the current ordering of the plurality of objects, the obtained ordering of the plurality of objects does not change, and the sort determining means stops the repeating operation.
  • the ranking determining means may calculate the current ordering of the plurality of objects according to the following formula:
  • R t is the current ordering of the plurality of objects
  • R t-1 is the ordering of the plurality of objects determined last time
  • l(p i , p j ) is the transition probability of object i to object j, where 1 ⁇ i ⁇ N, 1 ⁇ j ⁇ N, l (p i , p j ) can be calculated according to the transfer situation information, as calculated according to the following formula:
  • b ij represents the number of transitions from object i to object j
  • the sum of the probabilities of transferring from one object to all of the plurality of objects is 1, that is, l(p i , p j ) satisfies the following formula:
  • the initial ordered state matrix R 0 of the plurality of objects is as follows:
  • PageRank(p i ) 0 represents an initial sorting value of the object i
  • the PageRank(p i ) 0 can be manually set or can be determined according to the transition situation information.
  • the PageRank(p i ) 0 can be determined according to the following formula:
  • the above examples are only for better explaining the technical solutions of the present invention, and are not intended to limit the present invention.
  • the ordering of the objects, determining the implementation of the current ordering of the plurality of objects should be included in the scope of the present invention; for example, in the above formula for calculating R t , the offset matrix may adopt other values. , such as other manually set values, etc.; for example, if all l(p i , p j ) are not zero, then the above formula for calculating R t may not require the offset matrix, ie, the formula may tweak to:
  • the benchmark object determining means determines a benchmark object for at least one of the plurality of objects based on the ordering of the plurality of objects.
  • the benchmark object determining means selects the top Z objects among the plurality of objects as the common benchmark object of the objects after the plurality of objects.
  • Z is a preset natural number
  • the benchmark object determining device uses the highest ranked object as a common benchmark object for all other objects in the plurality of objects.
  • the target object determining means sets the object whose sorting position is close to the sorting position of the object as the target object of the object or the like.
  • the computer can only assist in manually determining the target object of the object by sorting the data from the object perspective, such as the industry and region to which the object belongs, the sales situation of the object, and the market demand.
  • the present invention provides a solution for a computer device to automatically determine a benchmark object; and, the present invention breaks the bias of those skilled in the art and can determine a benchmark object by analyzing the user's transitions among multiple objects.
  • the solution of the benchmark object is determined, and the user's perspective is more closely reflected, and the user's understanding of the benchmark object is more intuitively reflected. Therefore, the benchmark object determined by the solution of the present invention is more
  • the transfer situation information of the present invention is not common data. In fact, if the data from the user's point of view is explicitly mentioned, those skilled in the art are more likely to think of it. It is a direct evaluation from the user (such as scoring, commenting, etc.).
  • the present invention can be implemented in software and/or a combination of software and hardware.
  • the various devices of the present invention can be implemented using an application specific integrated circuit (ASIC) or any other similar hardware device.
  • the software program of the present invention may be executed by a processor to implement the steps or functions described above.
  • the software program (including related data structures) of the present invention can be stored in a computer readable recording medium such as a RAM memory, a magnetic or optical drive or a floppy disk and the like.
  • some of the steps or functions of the present invention may be implemented in hardware, for example, as a circuit that cooperates with a processor to perform various steps or functions.
  • CLAIMS 1. A method for determining a benchmark object for an object in a computer device, wherein the method comprises:
  • transfer situation information of the plurality of objects where the transfer situation information is used to indicate a transfer situation of the user in the plurality of objects based on the object information acquisition behavior;
  • a benchmark object is determined for at least one of the plurality of objects based on the ranking.
  • the last determined ranking is the ordering of the plurality of objects
  • the ranking of the plurality of objects determined last time is the initial ordering.
  • the ordering of the plurality of objects does not change.
  • step of obtaining transfer situation information between the keywords comprises:
  • Apparatus for determining a benchmark object for an object in a computer device comprising:
  • the last determined ranking is the ordering of the plurality of objects
  • the ranking of the plurality of objects determined last time is the initial ordering.
  • the ordering of the plurality of objects does not change.
  • Means for determining transfer situation information of the plurality of keywords according to the keyword attention record Means for determining transfer situation information of the plurality of keywords according to the keyword attention record.
  • a computer readable storage medium storing computer code, the method of any one of clauses 1 to 10 being executed when the computer code is executed.
  • a computer program product when the computer program product is executed by a computer device, the method of any one of clauses 1 to 10 being performed.
  • a computer device comprising a memory and a processor, the memory storing computer code, the processor being configured to execute the computer code to perform any of clauses 1 to 10 Said method.

Landscapes

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

Abstract

本发明设计用于在计算机设备中为对象确定标杆对象的方法,该方法包括:获取多个对象的转移情况信息;根据所述转移情况信息,为所述多个对象中的至少一个对象确定其标杆对象。本发明能够通过分析用户在多个对象中的转移情况,来确定标杆对象,因此,本发明的方案所确定的标杆对象更为客观、准确。

Description

一种用于确定标杆对象的方法和装置
相关申请的交叉引用
本申请享有2015年02月27日提交的专利申请号为201510090131.8、名称为“一种用于确定标杆对象的方法和装置”的中国专利申请的优先权,该在先申请的内容以引用方式合并于此。
技术领域
本发明涉及计算机技术领域,尤其涉及一种用于确定标杆对象的方法和装置。
背景技术
现有技术中,通常需要根据人工主观经验,来确定一个对象的标杆对象。在此过程中,计算机仅能通过整理从对象角度出发的数据,如该对象所属的行业和地域、对象的销售情况和市场需求等,来辅助人工判断对象的标杆对象。
发明内容
本发明的目的包括提供一种用于确定标杆对象的方法和装置。
根据本发明的一个方面,提供一种用于在计算机设备中为对象确定标杆对象的方法,其中,该方法包括:
获取多个对象的转移情况信息,所述转移情况信息用于指示基于对象信息获取行为的、用户在所述多个对象中的转移情况;
根据所述转移情况信息,为所述多个对象中的至少一个对象确定其标杆对象,其中,所述标杆对象属于所述多个对象。
根据本发明的另一个方面,还提供了一种用于在计算机设备中为对象确定标杆对象的装置,其中,该装置包括:
用于获取多个对象的转移情况信息的装置,所述转移情况信息用于指示基于对象信息获取行为的、用户在所述多个对象中的转移情况;
用于根据所述转移情况信息,为所述多个对象中的至少一个对象确定其标杆对象的装置,其中,所述标杆对象属于所述多个对象。
与现有技术相比,本发明具有以下优点:1)提供了计算机设备自动确定标杆对象的方案;2)打破了本领域技术人员的偏见,能够通过分析用户在多个对象中的转移情况,来确定标杆对象;3)相比从对象角度出发的数据,本发明中通过分析用户在多个对象中的转移情况来确定标杆对象的方案,更贴近用户角度,更能直观地反映用户对标杆对象的认识,因此,本发明的方案所确定的标杆对象更为客观、准确;4)即便在从用户角度出发的数据中,本发明的转移情况信息也并非常见数据,事实上,若明确提及从用户角度出发的数据,本领域技术人员更容易想到的是来自用户的直接评价(如打分、评论文字等)。
附图说明
通过阅读参照以下附图所作的对非限制性实施例所作的详细描述,本发明的其它特征、目的和优点将会变得更明显:
图1为本发明一个优选实施例的用于确定标杆对象的方法流程示意图;
图2为本发明一个优选实施例的用于确定标杆对象的确定装置的结构示意图;
图3示出了一个优选实施例的用户在多个对象中的转移路径的示意图;
图4示出了一个优选实施例的用户在多个关键词中的转移路径的示意图;
图5示出了一个从关键词的网状结构的转移路径到对象的网状结构的转移路径的转换的具体实例;
图6示出了一个优选实施例的、从一个对象到多个对象的转移的示意图;
图7示出了图6的一个具体实例。
附图中相同或相似的附图标记代表相同或相似的部件。
具体实施方式
下面结合附图对本发明作进一步详细描述。
图1为本发明一个优选实施例的确定标杆对象的方法流程示意图。其中,本实施例的方法主要通过计算机设备来实现;该计算机设备包括网络设备和用户设备。其中,所述网络设备包括但不限于单个网络服务器、多个网络服务器组成的服务器组或基于云计算(Cloud Computing)的由大量计算机或网络服务器构成的云,其中,云计算是分布式计算的一种,由一群松散耦合的计算机集组成的一个超级虚拟计算机;所述网络设备所处的网络包括但不限于互联网、广域网、城域网、局域网、VPN网络等。所述用户设备包括但不限于PC机、平板电脑、智能手机、PDA、IPTV等。
需要说明的是,所述计算机设备和网络仅为举例,其他现有的或今后可能出现的计算设备或网络如可适用于本发明,也应包含在本发明保护范围以内,并以引用方式包含于此。
根据本实施例的方法包括步骤S1和步骤S2。
在步骤S1中,计算机设备获取多个对象的转移情况信息。
优选地,所述对象具有商业性质。更优选地,所述对象包括品牌。
其中,所述转移情况信息用于指示基于对象信息获取行为的、用户在所述多个对象中的转移情况。其中,所述对象信息获取行为包括任何能够用于获取对象的信息的行为;例如,所述对象信息获取行为包括通过搜索与对象有关的关键词来获取对象信息的行为;又例如,所述对象信息获取行为包括通过点击并浏览与对象有关的内容来获取对象信息的行为。其中,所述“基于对象信息获取行为”,表示所述转移情况反映了用户在对象信息获取行为中产生的转移情况,优选地,所述转移情况需要基于对象信息获取行为来被确定;例如,通过统计多个用户在搜索行为中变更的搜索对象,或通过统计多个用户在搜索行为中变更的与对象关联的搜索关键词,来确定用户在中对象中 的转移情况信息等。
优选地,所述多个对象的转移情况信息包括但不限于以下至少一项:
1)用户在多个对象中的转移路径信息。
其中,所述转移路径信息指示了用户在多个对象中的转移路径。例如,存在三个对象Object1、Object2和Object3,转移路径信息指示多个用户在该三个对象中的转移路径包括:从Object1转移至Object2,以及,从Object1转移至Object3。
2)用户在各个对象之间的转移次数信息。
其中,所述转移次数信息指示了用户在各个对象之间的转移次数。例如,存在三个对象Object1、Object2和Object3,转移次数信息指示了多个用户在该三个对象之间的转移次数包括:从Object1转移至Object2五次,以及,从Object1转移至Object3八次。
3)用户在各个对象之间的转移概率信息。
其中,所述转移概率信息指示了用户在各个对象之间的转移概率。例如,存在三个对象Object1、Object2和Object3,转移概率信息指示了多个用户在该三个对象之间的转移概率包括:从Object1转移至Object2的概率为38.46%,以及,从Object1转移至Object3的概率为61.54%。
需要说明的是,多个对象中的部分对象之间可能不存在转移路径(即用户在对象信息获取行为中未在该等部分对象之间进行过转移),则该等部分对象之间的转移次数和转移概率均为零。此外,可能存在从一个对象转移至该对象自身的情况;例如,用户在搜索行为中可能连续多次采用不同的搜索关键词搜索同一个对象的信息,从而产生从一个对象转移至该对象自身的情况。
优选地,该转移情况信息可采用多种存储方式。
例如,该转移情况信息存储为表格,且表格中记录了用户在多个对象中的转移路径以及用户在各个对象之间的转移次数和转移概率,如下表1所示。
转移路径 转移次数 转移概率
Object1→Object2 5 38.46%
Object1→Object3 8 61.54%
表1
又例如,该转移情况信息包括:存储为网状结构的转移路径,以及,该网状结构中各个节点之间(即各个对象之间)的转移次数和/或转移概率。如对于9个对象Object1至Object9,该9个对象的转移情况信息包括如图3所示的转移路径,以及,图3中具有箭头连接的各个节点之间(如从Object1到Object2、从Object1到Object3、从Object1到Object4等等)的转移次数和/或转移概率。
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何用于指示基于对象信息获取行为的、用户在所述多个对象中的转移情况的转移情况信息,均应包含在本发明的范围内。
具体地,计算机设备获取多个对象的转移情况信息的方式包括但不限于:
1)计算机设备直接获取已预确定的、该多个对象的转移情况信息。
例如,计算机设备从本地或其他设备处读取已预确定的、该多个对象的转移情况信息。
2)步骤S1进一步包括步骤S11和步骤S12。
在步骤S11中,计算机设备获取多个关键词的转移情况信息。
其中,所述多个关键词的转移情况信息用于指示基于对象信息获取行为的、用户在多个关键词中的转移情况。优选地,所述多个关键词与对象信息获取行为中的对象相关联;例如,若对象信息获取行为为对象搜索行为,则关键词可为该搜索行为中用户输入的或选择的搜索关键词等。
优选地,所述多个关键词的转移情况信息包括以下至少一项:
a)用户在所述多个关键词中的转移路径信息。
其中,用户在所述多个关键词中的转移路径信息指示了用户在多个关键词中的转移路径。例如,存在三个关键词Query1、Query2和 Query3,转移路径信息指示多个用户在该三个关键词中的转移路径包括:从Query1转移至Query2,以及,从Query1转移至Query3。
b)用户在各个关键词之间的转移次数信息。
其中,用户在各个关键词之间的转移次数信息指示了用户在各个关键词之间的转移次数。例如,存在三个关键词Query1、Query2和Query3,转移次数信息指示多个用户在该三个关键词之间的转移次数包括:从Query1转移至Query2五次,以及,从Query1转移至Query3八次。
需要说明的是,多个关键词中的部分关键词之间可能不存在转移路径(即用户在对象信息获取行为中没有在该等部分关键词之间进行过转移),则该等部分关键词之间的转移次数为零。
优选地,该多个关键词的转移情况信息可采用多种存储方式。
例如,该转移情况信息存储为表格,且表格中记录了用户在多个关键词中的转移路径以及用户在各个关键词之间的转移次数,如下表2所示。
转移路径 转移次数
Query1→Query2 5
Query1→Query3 8
表2
又例如,该转移情况信息包括:存储为网状结构的转移路径,以及,该网状结构中各个节点之间(即各个关键词之间)的转移次数。如对于9个关键词Query1至Query9,该9个对象的转移情况信息包括如图4所示的转移路径,以及,图4中具有箭头连接的各个节点之间的转移次数。
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何用于指示基于对象信息获取行为的、用户在多个关键词中的转移情况的转移情况信息,均应包含在本发明的范围内。
具体地,计算机设备获取多个关键词的转移情况信息的实现方式包 括但不限于:
a)计算机设备直接获取已预确定的、该多个关键词的转移情况信息。
例如,计算机设备从本地或其他设备处读取已预确定的、该多个关键词的转移情况信息。
b)计算机设备获取至少一个用户的关键词关注记录,并根据所述关键词关注记录,确定所述多个关键词的转移情况信息。
其中,该关键词关注记录包括所述多个用户在对象信息获取行为中关注过的关键词以及所述关键词被关注的时间信息。优选地,对象信息获取行为包括搜索行为,所述关注过的关键词包括被搜索的关键词;优选地,对象信息获取行为包括浏览行为,所述关注过的关键词包括被点击来用以浏览对象内容的关键词。
优选地,对于每个用户的关键词关注记录,计算机设备根据该关键词关注记录中包含的关键词被关注的时间信息,确定该用户在关键词中的转移路径以及各个关键词之间的转移次数;并且,计算机设备通过合并各个用户在关键词中的转移路径以及各个关键词之间的转移次数,确定所述多个关键词的转移情况信息。
例如,计算机设备获取了用户A和用户B的关键词关注记录;其中,用户A和用户B的关键词关注记录分别如下表3和表4所示:
关注的关键词 关键词被关注的时间
Query1 2014-12-13-10:40
Query3 2014-12-13-10:36
表3
关注的关键词 关键词被关注的时间
Query1 2014-11-10-00:14
Query2 2014-11-10-00:23
表4
则对于用户A的关键词关注记录,计算机设备确定用户A在关键词 中的转移路径包括“Query1→Query3”,且“Query1→Query3”的转移次数为1;类似地,计算机设备确定用户B在关键词中的转移路径包括“Query1→Query2”,且“Query1→Query2”的转移次数为1。接着,计算机设备合并用户A和B在关键词中的转移路径以及各个关键词之间的转移次数,确定所述多个关键词的转移情况信息如下表5所示。
转移路径 转移次数
Query1→Query2 1
Query1→Query3 1
表5
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何获取多个关键词的转移情况信息的实现方式,均应包含在本发明的范围内。
在步骤S12中,计算机设备根据多个关键词分别关联至的对象以及所述多个关键词的转移情况信息,确定多个对象的转移情况信息。
具体地,计算机设备可根据用户在多个关键词中的转移路径信息以及多个关键词分别关联至的对象,确定用户在多个对象中的转移路径信息,并且,计算机设备根据用户在各个关键词之间的转移次数信息以及多个关键词分别关联至的对象,确定用户在各个对象之间的转移次数和/或转移概率信息。
例如,多个关键词的转移情况信息如前述表2所示,且Query1、Query2和Query3分别关联至Object1、Object2和Object3;计算机设备根据表2所示的关键词的转移情况信息以及前述关联关系,确定用户在对象Object1、Object2和Object3中的转移路径包括“Object1→Object2”和“Object1→Object3”,且该2个转移路径的转移次数分别为5和8;接着,计算机设备根据该2个转移路径的转移次数,计算“Object1→Object2”的转移概率=5/(5+8)=38.46%,“Object1→Object3”的转移概率=8/(5+8)=61.54%,也即,计算机设备获得如表1所示的转移情况信息。
需要说明的是,由于一个用户可能在连续多次对象信息获取行为中关注关联至同一个对象的不同关键词(如在多次搜索中采用对应同一个对象的不同搜素关键词),因此,可能存在从一个对象转移至该对象自身的转移概率,如可存在图6中所示的p00等。图6的一个具体实例可参见图7。如图7中所示,从“金宝贝”转移到“金宝贝”自身的概率可高达71.86%。
需要说明的是,优选地,计算机设备可基于以下公式来计算对象i到对象j的转移概率pij
Figure PCTCN2015092881-appb-000001
其中,aij表示对象i到对象j的转移次数,
Figure PCTCN2015092881-appb-000002
表示对象i到所有对象的转移次数。
例如,如图6所示,对象Object0可转移至其自身以及其他多个对象Object1至Object13;以从对象Object1转移到Object8为例,对象Objectp0到对象Object8的转移概率
Figure PCTCN2015092881-appb-000003
其中,
Figure PCTCN2015092881-appb-000004
表示从对象Object0到Object0自身以及对象Object1至Object13的所有转移次数。
需要说明的是,多个关键词分别关联至的对象可预先被确定,图5示出了从关键词的网状结构的转移路径到对象的网状结构的转移路径的转换实例。图5中,上方的网状结构中的各个节点为关键词,下方的网状结构中的各个节点为与上方的相应节点中的关键词对应的对象。
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何获取多个对象的转移情况信息的实现方式,均应包含在本发明的范围内。
在步骤S2中,计算机设备根据在步骤S1中获得的多个对象的转移情况信息,为多个对象中的至少一个对象确定其标杆对象,其中,所述标杆对象属于所述多个对象。优选地,所述多个对象属于同一类 别。
其中,一个对象的标杆(benchmark)对象,表示对该对象具有一定竞争力的对象,例如,在商业上与该对象产生较大竞争关系的对象等。
具体地,计算机设备根据多个对象的转移情况信息,为多个对象中的至少一个对象确定其标杆对象的实现方式包括但不限于:
1)计算机设备直接根据多个对象的转移情况信息,为多个对象中的至少一个对象确定其标杆对象。
例如,共存在三个对象Object1、Object2和Object3,转移情况信息如表1所示,则计算机设备根据从Object1到Object3的转移概率超过预定阈值60%,将Object3确定为Object1的标杆对象。
又例如,对于多个对象中的每个对象,计算机设备计算该多个对象中的其他对象到该对象的转移概率的平均值;接着,计算机设备选择平均值最高的一个对象,作为该多个对象中其他对象的共同的标杆对象。
需要说明的是,优选地,对于一个对象:从该对象转移到其他对象的转移次数或转移概率越高,则该其他对象被确定为该对象的标杆对象的可能性越高;从多个其他对象转移到该对象的次数总和、次数平均值、概率总和、和/或概率平均值越高,则该对象被确定为该多个其他对象的标杆对象的可能性越高。
2)步骤S2进一步包括步骤S21和步骤S22。
在步骤S21中,计算机设备根据多个对象的转移情况信息,确定该多个对象的排序。
其中,所述多个对象的排序可包括该多个对象中的每个对象在该多个对象中的排序位置,也可仅包括该多个对象中的部分对象中的每个的排序位置。
具体地,计算机设备根据多个对象的转移情况信息,确定该多个对象的排序的实现方式包括但不限于:
a)计算机设备直接根据多个对象的转移情况信息,确定该多个对象 的排序。
例如,对于多个对象中的每个对象,计算机设备计算该多个对象中的其他对象到该对象的转移概率的平均值;接着,计算机设备根据每个对象的所述转移概率的平均值,对多个对象进行排序。
需要说明的是,优选地,对于一个对象:从该对象转移到其他对象的转移概率越高,则该对象的排序较低的可能性越高;从多个其他对象转移到该对象的次数总和、次数平均值、概率总和、和/或概率平均值越高,则该对象排序越高的可能性越高。
b)计算机设备根据多个对象的转移情况信息以及所述多个对象的初始排序,确定所述多个对象的排序。其中,所述初始排序可人工设定,或通过多个对象的转移情况信息来被预确定。
具体地,计算机设备可根据多个对象的转移情况信息,对所述多个对象的初始排序进行调整,获得所述多个对象的排序。该实现方式包括但不限于:
i)计算机设备直接根据多个对象的转移情况信息,对所述多个对象的初始排序进行一次性调整,获得所述多个对象的排序。
例如,计算机设备根据转移情况信息中包含的转移次数和/或转移概率,将转移到其他对象的概率较低,且其他对象转移到自身的概率较高的对象的排序上调等。
ii)计算机设备重复执行以下步骤,直至满足预定条件:计算机设备根据多个对象的转移情况信息以及上一次确定的所述多个对象的排序,确定所述多个对象的本次的排序;更优选地,计算机设备可根据对象的转移情况信息的矩阵与上一次确定的多个对象的排序的矩阵,如通过将该两个矩阵相乘,来确定多个对象的本次的排序。其中,最后一次被确定的排序为所述多个对象的排序;其中,当确定所述多个对象的本次的排序的步骤被首次执行时,所述上一次确定的所述多个对象的排序为所述初始排序。
其中,所述预定条件包括任何能够终止所述重复步骤的条件。优选地,所述预定条件包括但不限于以下至少一项:
-重复次数达到预定重复阈值;例如,预定重复阈值为50次,当计算机设备执行50次根据多个对象的转移情况信息以及上一次确定的所述多个对象的排序来确定所述多个对象的本次的排序的操作后,停止重复操作。
-在多次重复所述确定所述多个对象的本次的排序的步骤后,所述多个对象的排序没有发生变化;例如,当计算机设备重复执行5次根据多个对象的转移情况信息以及上一次确定的所述多个对象的排序来确定所述多个对象的本次的排序后,所获得的该多个对象的排序没有发生变化,则计算机设备停止重复操作。
优选地,在实现方式ii)中,计算机设备可根据以下公式来计算多个对象的本次的排序:
Figure PCTCN2015092881-appb-000005
其中,Rt为多个对象的本次的排序;Rt-1为上一次确定的多个对象的排序;
Figure PCTCN2015092881-appb-000006
为偏移量矩阵,其中,q为参数,且q可预先设定,N为多个对象的数量;l(pi,pj)为对象i到对象j的转移概率,其中,1≤i≤N,1≤j≤N,l(pi,pj)可根据转移情况信息计算,如可根据以下公式计算:
Figure PCTCN2015092881-appb-000007
其中,bij表示从对象i到对象j的转移次数,
Figure PCTCN2015092881-appb-000008
表示从对象i到所述多个对象中的所有对象的转移次数。此外,从一个对象转移到所述多个对象中的所有对象的概率之和为1,也即,l(pi,pj)满足以下公式:
Figure PCTCN2015092881-appb-000009
优选地,当t=1时(即第一次执行确定多个对象的排序的步骤时),多个对象的初始排序的状态矩阵R0如下所示:
Figure PCTCN2015092881-appb-000010
其中,PageRank(pi)0表示对象i的初始排序值,该PageRank(pi)0可人工设定,或可根据转移情况信息确定。例如,该PageRank(pi)0可根据如下公式确定:
Figure PCTCN2015092881-appb-000011
其中,ai表示对象i在转移次数信息中出现的次数,例如,从对象i转移到其他所有对象的次数为301次,从其他所有对象转移到对象i的次数为298次,则对象i在转移次数信息中出现的次数=301+298=599次;
Figure PCTCN2015092881-appb-000012
表示所有品牌在转移次数信息中出现的次数总和。
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何根据多个对象的转移情况信息以及上一次确定的所述多个对象的排序,确定所述多个对象的本次的排序的实现方式,均应包含在本发明的范围内;例如,上述用于计算Rt的公式中,偏移量矩阵可采用其他数值,如其他人工设定的值等;又例如,若所有l(pi,pj)均不为零,则上述用于计算Rt的公式可无需所述偏移量矩阵,即该公式可调整为:
Figure PCTCN2015092881-appb-000013
或者,对于为零的l(pi,pj),若计算机设备赋予其一个极小的非零值,则同样可采用上述调整后的公式。
在步骤S22中,计算机设备根据多个对象的排序,为所述多个对象中的至少一个对象确定其标杆对象。
例如,计算机设备将多个对象中排序最高的前Z个对象作为该多个对象排序在Z之后的对象的共同的标杆对象。其中,Z为预先设定的自然数,当Z为1时,计算机设备将排序最高的一个对象作为多个对象中其他所有对象的共同的标杆对象。
又例如,对于多个对象中的一个对象,计算机设备将排序位置与该对象的排序位置接近的对象,作为该对象的标杆对象等。
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何根据多个对象的排序,为所述多个对象中的至少一个对象确定其标杆对象的实现方式,均应包含在本发明的范围内。
现有技术中,通常需要根据人工主观经验,来确定一个对象的标杆对象。在此过程中,计算机仅能通过整理从对象角度出发的数据,如该对象所属的行业和地域、对象的销售情况和市场需求等,来辅助人工判断对象的标杆对象。
也即,现有技术中尚不存在可由计算机独立确定标杆对象的方案;此外,受到目前人工主观经验的影响,本领域技术人员存在这样的偏见:在考虑标杆对象的确定方案时,往往仅关注前述从对象角度出发的数据。
本发明提供了计算机设备自动确定标杆对象的方案;并且,本发明打破了本领域技术人员的偏见,能够通过分析用户在多个对象中的转移情况,来确定标杆对象。此外,相比前述从对象角度出发的数据,本发明中通过分析用户在多个对象中的转移情况来确定标杆对象的方案,更贴近用户角度,更能直观地反映用户对标杆对象的认识,因此,本发明的方案所确定的标杆对象更为客观、准确;此外,即便在从用户角度出发的数据中,本发明的转移情况信息也并非常见数据, 事实上,若明确提及从用户角度出发的数据,本领域技术人员更容易想到的是来自用户的直接评价(如打分、评论文字等)。
图2为本发明一个优选实施例的用于确定标杆对象的确定装置的结构示意图。该确定装置可安装至计算机设备中,该确定装置包括用于获取多个对象的转移情况信息的装置(以下简称“获取装置1”)、用于根据所述转移情况信息,为所述多个对象中的至少一个对象确定其标杆对象的装置(以下简称“子确定装置2”)。
获取装置1获取多个对象的转移情况信息。
优选地,所述对象具有商业性质。更优选地,所述对象包括品牌。
其中,所述转移情况信息用于指示基于对象信息获取行为的、用户在所述多个对象中的转移情况。其中,所述对象信息获取行为包括任何能够用于获取对象的信息的行为;例如,所述对象信息获取行为包括通过搜索与对象有关的关键词来获取对象信息的行为;又例如,所述对象信息获取行为包括通过点击并浏览与对象有关的内容来获取对象信息的行为。其中,所述“基于对象信息获取行为”,表示所述转移情况反映了用户在对象信息获取行为中产生的转移情况,优选地,所述转移情况需要基于对象信息获取行为来被确定;例如,通过统计多个用户在搜索行为中变更的搜索对象,或通过统计多个用户在搜索行为中变更的与对象关联的搜索关键词,来确定用户在中对象中的转移情况信息等。
优选地,所述多个对象的转移情况信息包括但不限于以下至少一项:
1)用户在多个对象中的转移路径信息。
其中,所述转移路径信息指示了用户在多个对象中的转移路径。例如,存在三个对象Object1、Object2和Object3,转移路径信息指示多个用户在该三个对象中的转移路径包括:从Object1转移至Object2,以及,从Object1转移至Object3。
2)用户在各个对象之间的转移次数信息。
其中,所述转移次数信息指示了用户在各个对象之间的转移次数。例如,存在三个对象Object1、Object2和Object3,转移次数信息指示了多个用户在该三个对象之间的转移次数包括:从Object1转移至Object2五次,以及,从Object1转移至Object3八次。
3)用户在各个对象之间的转移概率信息。
其中,所述转移概率信息指示了用户在各个对象之间的转移概率。例如,存在三个对象Object1、Object2和Object3,转移概率信息指示了多个用户在该三个对象之间的转移概率包括:从Object1转移至Object2的概率为38.46%,以及,从Object1转移至Object3的概率为61.54%。
需要说明的是,多个对象中的部分对象之间可能不存在转移路径(即用户在对象信息获取行为中未在该等部分对象之间进行过转移),则该等部分对象之间的转移次数和转移概率均为零。此外,可能存在从一个对象转移至该对象自身的情况;例如,用户在搜索行为中可能连续多次采用不同的搜索关键词搜索同一个对象的信息,从而产生从一个对象转移至该对象自身的情况。
优选地,该转移情况信息可采用多种存储方式。
例如,该转移情况信息存储为表格,且表格中记录了用户在多个对象中的转移路径以及用户在各个对象之间的转移次数和转移概率,如前述表1所示。
又例如,该转移情况信息包括:存储为网状结构的转移路径,以及,该网状结构中各个节点之间(即各个对象之间)的转移次数和/或转移概率。如对于9个对象Object1至Object9,该9个对象的转移情况信息包括如图3所示的转移路径,以及,图3中具有箭头连接的各个节点之间(如从Object1到Object2、从Object1到Object3、从Object1到Object4等等)的转移次数和/或转移概率。
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何用于指示基于对象信息获取行为的、用户在所述多个对象中的转移情况的转移情况信息,均应包含在本发明的范围内。
具体地,获取装置1获取多个对象的转移情况信息的方式包括但不限于:
1)获取装置1直接获取已预确定的、该多个对象的转移情况信息。
例如,获取装置1从本地或其他设备处读取已预确定的、该多个对象的转移情况信息。
2)获取装置1进一步包括用于获取多个关键词的转移情况信息的装置(以下简称“子获取装置”,图未示)和用于根据多个关键词分别关联至的对象以及所述多个关键词的转移情况信息,确定多个对象的转移情况信息的装置(以下简称“转移情况确定装置”,图未示)。
子获取装置获取多个关键词的转移情况信息。
其中,所述多个关键词的转移情况信息用于指示基于对象信息获取行为的、用户在多个关键词中的转移情况。优选地,所述多个关键词与对象信息获取行为中的对象相关联;例如,若对象信息获取行为为对象搜索行为,则关键词可为该搜索行为中用户输入的或选择的搜索关键词等。
优选地,所述多个关键词的转移情况信息包括以下至少一项:
a)用户在所述多个关键词中的转移路径信息。
其中,用户在所述多个关键词中的转移路径信息指示了用户在多个关键词中的转移路径。例如,存在三个关键词Query1、Query2和Query3,转移路径信息指示多个用户在该三个关键词中的转移路径包括:从Query1转移至Query2,以及,从Query1转移至Query3。
b)用户在各个关键词之间的转移次数信息。
其中,用户在各个关键词之间的转移次数信息指示了用户在各个关键词之间的转移次数。例如,存在三个关键词Query1、Query2和Query3,转移次数信息指示多个用户在该三个关键词之间的转移次数包括:从Query1转移至Query2五次,以及,从Query1转移至Query3八次。
需要说明的是,多个关键词中的部分关键词之间可能不存在转移路径(即用户在对象信息获取行为中没有在该等部分关键词之间进行过转 移),则该等部分关键词之间的转移次数为零。
优选地,该多个关键词的转移情况信息可采用多种存储方式。
例如,该转移情况信息存储为表格,且表格中记录了用户在多个关键词中的转移路径以及用户在各个关键词之间的转移次数,如前述表2所示。
又例如,该转移情况信息包括:存储为网状结构的转移路径,以及,该网状结构中各个节点之间(即各个关键词之间)的转移次数。如对于9个关键词Query1至Query9,该9个对象的转移情况信息包括如图4所示的转移路径,以及,图4中具有箭头连接的各个节点之间的转移次数。
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何用于指示基于对象信息获取行为的、用户在多个关键词中的转移情况的转移情况信息,均应包含在本发明的范围内。
具体地,子获取装置获取多个关键词的转移情况信息的实现方式包括但不限于:
a)子获取装置直接获取已预确定的、该多个关键词的转移情况信息。
例如,子获取装置从本地或其他设备处读取已预确定的、该多个关键词的转移情况信息。
b)子获取装置获取至少一个用户的关键词关注记录,并根据所述关键词关注记录,确定所述多个关键词的转移情况信息。
其中,该关键词关注记录包括所述多个用户在对象信息获取行为中关注过的关键词以及所述关键词被关注的时间信息。优选地,对象信息获取行为包括搜索行为,所述关注过的关键词包括被搜索的关键词;优选地,对象信息获取行为包括浏览行为,所述关注过的关键词包括被点击来用以浏览对象内容的关键词。
优选地,对于每个用户的关键词关注记录,子获取装置根据该关键词关注记录中包含的关键词被关注的时间信息,确定该用户在关键词 中的转移路径以及各个关键词之间的转移次数;并且,子获取装置通过合并各个用户在关键词中的转移路径以及各个关键词之间的转移次数,确定所述多个关键词的转移情况信息。
例如,子获取装置获取了用户A和用户B的关键词关注记录;其中,用户A和用户B的关键词关注记录分别如前述表3和表4所示。
则对于用户A的关键词关注记录,子获取装置确定用户A在关键词中的转移路径包括“Query1→Query3”,且“Query1→Query3”的转移次数为1;类似地,子获取装置确定用户B在关键词中的转移路径包括“Query1→Query2”,且“Query1→Query2”的转移次数为1。接着,子获取装置合并用户A和B在关键词中的转移路径以及各个关键词之间的转移次数,确定所述多个关键词的转移情况信息如前述表5所示。
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何获取多个关键词的转移情况信息的实现方式,均应包含在本发明的范围内。
转移情况确定装置根据多个关键词分别关联至的对象以及所述多个关键词的转移情况信息,确定多个对象的转移情况信息。
具体地,转移情况确定装置可根据用户在多个关键词中的转移路径信息以及多个关键词分别关联至的对象,确定用户在多个对象中的转移路径信息,并且,转移情况确定装置根据用户在各个关键词之间的转移次数信息以及多个关键词分别关联至的对象,确定用户在各个对象之间的转移次数和/或转移概率信息。
例如,多个关键词的转移情况信息如前述表2所示,且Query1、Query2和Query3分别关联至Object1、Object2和Object3;转移情况确定装置根据表2所示的关键词的转移情况信息以及前述关联关系,确定用户在对象Object1、Object2和Object3中的转移路径包括“Object1→Object2”和“Object1→Object3”,且该2个转移路径的转移次数分别为5和8;接着,转移情况确定装置根据该2个转移路径的转移次数,计算“Object1→Object2”的转移概率=5/(5+8)=38.46%,“Object1→Object3”的转移概率=8/(5+8)=61.54%,也即,转移情况确定装置获得 如表1所示的转移情况信息。
需要说明的是,由于一个用户可能在连续多次对象信息获取行为中关注关联至同一个对象的不同关键词(如在多次搜索中采用对应同一个对象的不同搜素关键词),因此,可能存在从一个对象转移至该对象自身的转移概率,如可存在图6中所示的p00等。图6的一个具体实例可参见图7。如图7中所示,从“金宝贝”转移到“金宝贝”自身的概率可高达71.86%。
需要说明的是,优选地,转移情况确定装置可基于以下公式来计算对象i到对象j的转移概率pij
Figure PCTCN2015092881-appb-000014
其中,aij表示对象i到对象j的转移次数,
Figure PCTCN2015092881-appb-000015
表示对象i到所有对象的转移次数。
例如,如图6所示,对象Object0可转移至其自身以及其他多个对象Object1至Object13;以从对象Object1转移到Object8为例,对象Objectp0到对象Object8的转移概率
Figure PCTCN2015092881-appb-000016
其中,
Figure PCTCN2015092881-appb-000017
表示从对象Object0到Object0自身以及对象Object1至Object13的所有转移次数。
需要说明的是,多个关键词分别关联至的对象可预先被确定,图5示出了从关键词的网状结构的转移路径到对象的网状结构的转移路径的转换实例。图5中,上方的网状结构中的各个节点为关键词,下方的网状结构中的各个节点为与上方的相应节点中的关键词对应的对象。
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何获取多个对象的转移情况信息的实现方式,均应包含在本发明的范围内。
子确定装置2根据获取装置1获得的多个对象的转移情况信息,为多个对象中的至少一个对象确定其标杆对象,其中,所述标杆对象属 于所述多个对象。优选地,所述多个对象属于同一类别。
其中,一个对象的标杆(benchmark)对象,表示对该对象具有一定竞争力的对象,例如,在商业上与该对象产生较大竞争关系的对象等。
具体地,子确定装置2根据多个对象的转移情况信息,为多个对象中的至少一个对象确定其标杆对象的实现方式包括但不限于:
1)子确定装置2直接根据多个对象的转移情况信息,为多个对象中的至少一个对象确定其标杆对象。
例如,共存在三个对象Object1、Object2和Object3,转移情况信息如表1所示,则子确定装置2根据从Object1到Object3的转移概率超过预定阈值60%,将Object3确定为Object1的标杆对象。
又例如,对于多个对象中的每个对象,子确定装置2计算该多个对象中的其他对象到该对象的转移概率的平均值;接着,子确定装置2选择平均值最高的一个对象,作为该多个对象中其他对象的共同的标杆对象。
需要说明的是,优选地,对于一个对象:从该对象转移到其他对象的转移次数或转移概率越高,则该其他对象被确定为该对象的标杆对象的可能性越高;从多个其他对象转移到该对象的次数总和、次数平均值、概率总和、和/或概率平均值越高,则该对象被确定为该多个其他对象的标杆对象的可能性越高。
2)子确定装置2进一步包括用于根据所述转移情况信息,确定所述多个对象的排序的装置(以下简称“排序确定装置”,图未示)、用于根据所述排序,为所述多个对象中的至少一个对象确定其标杆对象的装置(以下简称“标杆对象确定装置”,图未示)。
排序确定装置根据多个对象的转移情况信息,确定该多个对象的排序。
其中,所述多个对象的排序可包括该多个对象中的每个对象在该多个对象中的排序位置,也可仅包括该多个对象中的部分对象中的每个的排序位置。
具体地,排序确定装置根据多个对象的转移情况信息,确定该多个对象的排序的实现方式包括但不限于:
a)排序确定装置直接根据多个对象的转移情况信息,确定该多个对象的排序。
例如,对于多个对象中的每个对象,排序确定装置计算该多个对象中的其他对象到该对象的转移概率的平均值;接着,排序确定装置根据每个对象的所述转移概率的平均值,对多个对象进行排序。
需要说明的是,优选地,对于一个对象:从该对象转移到其他对象的转移概率越高,则该对象的排序较低的可能性越高;从多个其他对象转移到该对象的次数总和、次数平均值、概率总和、和/或概率平均值越高,则该对象排序越高的可能性越高。
b)排序确定装置根据多个对象的转移情况信息以及所述多个对象的初始排序,确定所述多个对象的排序。其中,所述初始排序可人工设定,或通过多个对象的转移情况信息来被预确定。
具体地,排序确定装置可根据多个对象的转移情况信息,对所述多个对象的初始排序进行调整,获得所述多个对象的排序。该实现方式包括但不限于:
i)排序确定装置直接根据多个对象的转移情况信息,对所述多个对象的初始排序进行一次性调整,获得所述多个对象的排序。
例如,排序确定装置根据转移情况信息中包含的转移次数和/或转移概率,将转移到其他对象的概率较低,且其他对象转移到自身的概率较高的对象的排序上调等。
ii)排序确定装置重复执行以下操作,直至满足预定条件:排序确定装置根据多个对象的转移情况信息以及上一次确定的所述多个对象的排序,确定所述多个对象的本次的排序;更优选地,排序确定装置可根据对象的转移情况信息的矩阵与上一次确定的多个对象的排序的矩阵,如通过将该两个矩阵相乘,来确定多个对象的本次的排序。其中,最后一次被确定的排序为所述多个对象的排序;其中,当确定所述多个对象的本次的排序的操作被首次执行时,所述上一次确定的所述多 个对象的排序为所述初始排序。
其中,所述预定条件包括任何能够终止所述重复操作的条件。优选地,所述预定条件包括但不限于以下至少一项:
-重复次数达到预定重复阈值;例如,预定重复阈值为50次,当排序确定装置执行50次根据多个对象的转移情况信息以及上一次确定的所述多个对象的排序来确定所述多个对象的本次的排序的操作后,停止重复操作。
-在多次重复所述确定所述多个对象的本次的排序的操作后,所述多个对象的排序没有发生变化;例如,当排序确定装置重复执行5次根据多个对象的转移情况信息以及上一次确定的所述多个对象的排序来确定所述多个对象的本次的排序后,所获得的该多个对象的排序没有发生变化,则排序确定装置停止重复操作。
优选地,在实现方式ii)中,排序确定装置可根据以下公式来计算多个对象的本次的排序:
Figure PCTCN2015092881-appb-000018
其中,Rt为多个对象的本次的排序;Rt-1为上一次确定的多个对象的排序;
Figure PCTCN2015092881-appb-000019
为偏移量矩阵,其中,q为参数,且q可预先设定,N为多个对象的数量;l(pi,pj)为对象i到对象j的转移概率,其中,1≤i≤N,1≤j≤N,l(pi,pj)可根据转移情况信息计算,如可根据以下公式计算:
Figure PCTCN2015092881-appb-000020
其中,bij表示从对象i到对象j的转移次数,
Figure PCTCN2015092881-appb-000021
表示从对象i 到所述多个对象中的所有对象的转移次数。此外,从一个对象转移到所述多个对象中的所有对象的概率之和为1,也即,l(pi,pj)满足以下公式:
Figure PCTCN2015092881-appb-000022
优选地,当t=1时(即第一次执行确定多个对象的排序的操作时),多个对象的初始排序的状态矩阵R0如下所示:
Figure PCTCN2015092881-appb-000023
其中,PageRank(pi)0表示对象i的初始排序值,该PageRank(pi)0可人工设定,或可根据转移情况信息确定。例如,该PageRank(pi)0可根据如下公式确定:
Figure PCTCN2015092881-appb-000024
其中,ai表示对象i在转移次数信息中出现的次数,例如,从对象i转移到其他所有对象的次数为301次,从其他所有对象转移到对象i的次数为298次,则对象i在转移次数信息中出现的次数=301+298=599次;
Figure PCTCN2015092881-appb-000025
表示所有品牌在转移次数信息中出现的次数总和。
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何根据多个对象的转移情况信息以及上一次确定的所述多个对象的排序,确定所述多个对象的本次的排序的实现方式,均应包含在本发明的范围内;例如,上述用于计算Rt的公式中,偏移量矩阵可采用其他数值,如其他人工设定的值等;又例如,若所有l(pi,pj)均不为零,则上述用于计算Rt的公式可无需所述偏移量矩阵,即该公式可调整为:
Figure PCTCN2015092881-appb-000026
或者,对于为零的l(pi,pj),若确定装置赋予其一个极小的非零值,则同样可采用上述调整后的公式。
标杆对象确定装置根据多个对象的排序,为所述多个对象中的至少一个对象确定其标杆对象。
例如,标杆对象确定装置将多个对象中排序最高的前Z个对象作为该多个对象排序在Z之后的对象的共同的标杆对象。其中,Z为预先设定的自然数,当Z为1时,标杆对象确定装置将排序最高的一个对象作为多个对象中其他所有对象的共同的标杆对象。
又例如,对于多个对象中的一个对象,标杆对象确定装置将排序位置与该对象的排序位置接近的对象,作为该对象的标杆对象等。
需要说明的是,上述举例仅为更好地说明本发明的技术方案,而非对本发明的限制,本领域技术人员应该理解,任何根据多个对象的排序,为所述多个对象中的至少一个对象确定其标杆对象的实现方式,均应包含在本发明的范围内。
现有技术中,通常需要根据人工主观经验,来确定一个对象的标杆对象。在此过程中,计算机仅能通过整理从对象角度出发的数据,如该对象所属的行业和地域、对象的销售情况和市场需求等,来辅助人工判断对象的标杆对象。
也即,现有技术中尚不存在可由计算机独立确定标杆对象的方案;此外,受到目前人工主观经验的影响,本领域技术人员存在这样的偏见:在考虑标杆对象的确定方案时,往往仅关注前述从对象角度出发的数据。
本发明提供了计算机设备自动确定标杆对象的方案;并且,本发明打破了本领域技术人员的偏见,能够通过分析用户在多个对象中的转移情况,来确定标杆对象。此外,相比前述从对象角度出发的数据, 本发明中通过分析用户在多个对象中的转移情况来确定标杆对象的方案,更贴近用户角度,更能直观地反映用户对标杆对象的认识,因此,本发明的方案所确定的标杆对象更为客观、准确;此外,即便在从用户角度出发的数据中,本发明的转移情况信息也并非常见数据,事实上,若明确提及从用户角度出发的数据,本领域技术人员更容易想到的是来自用户的直接评价(如打分、评论文字等)。
需要注意的是,本发明可在软件和/或软件与硬件的组合体中被实施,例如,本发明的各个装置可采用专用集成电路(ASIC)或任何其他类似硬件设备来实现。在一个实施例中,本发明的软件程序可以通过处理器执行以实现上文所述步骤或功能。同样地,本发明的软件程序(包括相关的数据结构)可以被存储到计算机可读记录介质中,例如,RAM存储器,磁或光驱动器或软磁盘及类似设备。另外,本发明的一些步骤或功能可采用硬件来实现,例如,作为与处理器配合从而执行各个步骤或功能的电路。
对于本领域技术人员而言,显然本发明不限于上述示范性实施例的细节,而且在不背离本发明的精神或基本特征的情况下,能够以其他的具体形式实现本发明。因此,无论从哪一点来看,均应将实施例看作是示范性的,而且是非限制性的,本发明的范围由所附权利要求而不是上述说明限定,因此旨在将落在权利要求的等同要件的含义和范围内的所有变化涵括在本发明内。不应将权利要求中的任何附图标记视为限制所涉及的权利要求。此外,显然“包括”一词不排除其他单元或步骤,单数不排除复数。系统权利要求中陈述的多个单元或装置也可以由一个单元或装置通过软件或者硬件来实现。第一,第二等词语用来表示名称,而并不表示任何特定的顺序。
在权利要求书中规定了各个实施例的各个方面。在下列编号条款中规定了各个实施例的这些和其他方面:
1.一种用于在计算机设备中为对象确定标杆对象的方法,其中,该方法包括:
获取多个对象的转移情况信息,所述转移情况信息用于指示基于对象信息获取行为的、用户在所述多个对象中的转移情况;
根据所述转移情况信息,为所述多个对象中的至少一个对象确定其标杆对象,其中,所述标杆对象属于所述多个对象。
2.根据条款1所述的方法,其中,确定所述标杆对象的步骤包括:
根据所述转移情况信息,确定所述多个对象的排序;
根据所述排序,为所述多个对象中的至少一个对象确定其标杆对象。
3.根据条款1或2所述的方法,其中,确定所述多个对象的排序的步骤包括:
根据所述转移情况信息以及所述多个对象的初始排序,确定所述多个对象的排序。
4.根据条款3所述的方法,其中,确定所述多个对象的排序的步骤包括:
重复执行以下步骤,直至满足预定条件:
根据所述转移情况信息以及上一次确定的所述多个对象的排序,确定所述多个对象的本次的排序;
其中,最后一次被确定的排序为所述多个对象的排序;
其中,当确定所述多个对象的本次的排序的步骤被首次执行时,所述上一次确定的所述多个对象的排序为所述初始排序。
5.根据条款4所述的方法,其中,所述预定条件包括以下至少一项:
-重复次数达到预定重复阈值;
-在多次重复所述确定所述多个对象的本次的排序的步骤后,所述多个对象的排序没有发生变化。
6.根据条款1至5中任一项所述的方法,其中,所述多个对象属于同一类别。
7.根据条款1至6中任一项所述的方法,其中,获取所述转移情况信息的步骤包括:
获取多个关键词的转移情况信息,其中,所述多个关键词的转移情况信息用于指示基于对象信息获取行为的、用户在多个关键词中的转移情况;
根据所述多个关键词分别关联至的对象以及所述多个关键词的转移情况信息,确定所述多个对象的转移情况信息。
8.根据条款7所述的方法,其中,获取所述关键词之间的转移情况信息的步骤包括:
获取至少一个用户的关键词关注记录,该关键词关注记录包括所述多个用户在对象信息获取行为中关注过的关键词以及所述关键词被关注的时间信息;
根据所述关键词关注记录,确定所述多个关键词的转移情况信息。
9.根据条款7或8所述的方法,其中,所述多个关键词的转移情况信息包括以下至少一项:
-用户在所述多个关键词中的转移路径信息;
-用户在各个关键词之间的转移次数信息。
10.根据条款1至9中任一项所述的方法,其中,所述多个对象之间的转移情况信息包括以下至少一项:
-用户在所述多个对象中的转移路径信息;
-用户在各个对象之间的转移次数信息;
-用户在各个对象之间的转移概率信息。
11.一种用于在计算机设备中为对象确定标杆对象的装置,其中,该装置包括:
用于获取多个对象的转移情况信息的装置,所述转移情况信息用于指示基于对象信息获取行为的、用户在所述多个对象中的转移情况;
用于根据所述转移情况信息,为所述多个对象中的至少一个对象确定其标杆对象的装置,其中,所述标杆对象属于所述多个对象。
12.根据条款11所述的装置,其中,用于确定所述标杆对象的装置包括:
用于根据所述转移情况信息,确定所述多个对象的排序的装置;
用于根据所述排序,为所述多个对象中的至少一个对象确定其标杆对象的装置。
13.根据条款11或12所述的装置,其中,用于确定所述多个对象的排序的装置包括:
用于根据所述转移情况信息以及所述多个对象的初始排序,确定所述多个对象的排序的装置。
14.根据条款13所述的装置,其中,用于确定所述多个对象的排序的装置包括:
用于触发以下装置重复执行操作,直至满足预定条件的装置:
用于根据所述转移情况信息以及上一次确定的所述多个对象的排序,确定所述多个对象的本次的排序的装置;
其中,最后一次被确定的排序为所述多个对象的排序;
其中,当确定所述多个对象的本次的排序的步骤被首次执行时,所述上一次确定的所述多个对象的排序为所述初始排序。
15.根据条款14所述的装置,其中,所述预定条件包括以下至少一项:
-重复次数达到预定重复阈值;
-在多次重复所述确定所述多个对象的本次的排序的步骤后,所述多个对象的排序没有发生变化。
16.根据条款11至15中任一项所述的装置,其中,所述多个对象属于同一类别。
17.根据条款11至16中任一项所述的装置,其中,用于获取所述转移情况信息的装置包括:
用于获取多个关键词的转移情况信息的装置,其中,所述多个关键词的转移情况信息用于指示基于对象信息获取行为的、用户在多个关键词中的转移情况;
用于根据所述多个关键词分别关联至的对象以及所述多个关键词的转移情况信息,确定所述多个对象的转移情况信息的装置。
18.根据条款17所述的装置,其中,获取所述关键词之间的转移情 况信息的装置包括:
用于获取至少一个用户的关键词关注记录的装置,该关键词关注记录包括所述多个用户在对象信息获取行为中关注过的关键词以及所述关键词被关注的时间信息;
用于根据所述关键词关注记录,确定所述多个关键词的转移情况信息的装置。
19.根据条款17或18所述的装置,其中,所述多个关键词的转移情况信息包括以下至少一项:
-用户在所述多个关键词中的转移路径信息;
-用户在各个关键词之间的转移次数信息。
20.根据条款11至19中任一项所述的装置,其中,所述多个对象之间的转移情况信息包括以下至少一项:
-用户在所述多个对象中的转移路径信息;
-用户在各个对象之间的转移次数信息;
-用户在各个对象之间的转移概率信息。
21.一种计算机可读存储介质,所述计算机可读存储介质存储有计算机代码,当所述计算机代码被执行时,如条款1至10中任一项所述的方法被执行。
22.一种计算机程序产品,当所述计算机程序产品被计算机设备执行时,如条款1至10中任一项所述的方法被执行。
23.一种计算机设备,所述计算机设备包括存储器和处理器,所述存储器中存储有计算机代码,所述处理器被配置来通过执行所述计算机代码以执行如条款1至10中任一项所述的方法。

Claims (23)

  1. 一种用于在计算机设备中为对象确定标杆对象的方法,其中,该方法包括:
    获取多个对象的转移情况信息,所述转移情况信息用于指示基于对象信息获取行为的、用户在所述多个对象中的转移情况;
    根据所述转移情况信息,为所述多个对象中的至少一个对象确定其标杆对象,其中,所述标杆对象属于所述多个对象。
  2. 根据权利要求1所述的方法,其中,确定所述标杆对象的步骤包括:
    根据所述转移情况信息,确定所述多个对象的排序;
    根据所述排序,为所述多个对象中的至少一个对象确定其标杆对象。
  3. 根据权利要求1或2所述的方法,其中,确定所述多个对象的排序的步骤包括:
    根据所述转移情况信息以及所述多个对象的初始排序,确定所述多个对象的排序。
  4. 根据权利要求3所述的方法,其中,确定所述多个对象的排序的步骤包括:
    重复执行以下步骤,直至满足预定条件:
    根据所述转移情况信息以及上一次确定的所述多个对象的排序,确定所述多个对象的本次的排序;
    其中,最后一次被确定的排序为所述多个对象的排序;
    其中,当确定所述多个对象的本次的排序的步骤被首次执行时,所述上一次确定的所述多个对象的排序为所述初始排序。
  5. 根据权利要求4所述的方法,其中,所述预定条件包括以下至少一项:
    -重复次数达到预定重复阈值;
    -在多次重复所述确定所述多个对象的本次的排序的步骤后,所述 多个对象的排序没有发生变化。
  6. 根据权利要求1至5中任一项所述的方法,其中,所述多个对象属于同一类别。
  7. 根据权利要求1至6中任一项所述的方法,其中,获取所述转移情况信息的步骤包括:
    获取多个关键词的转移情况信息,其中,所述多个关键词的转移情况信息用于指示基于对象信息获取行为的、用户在多个关键词中的转移情况;
    根据所述多个关键词分别关联至的对象以及所述多个关键词的转移情况信息,确定所述多个对象的转移情况信息。
  8. 根据权利要求7所述的方法,其中,获取所述关键词之间的转移情况信息的步骤包括:
    获取至少一个用户的关键词关注记录,该关键词关注记录包括所述多个用户在对象信息获取行为中关注过的关键词以及所述关键词被关注的时间信息;
    根据所述关键词关注记录,确定所述多个关键词的转移情况信息。
  9. 根据权利要求7或8所述的方法,其中,所述多个关键词的转移情况信息包括以下至少一项:
    -用户在所述多个关键词中的转移路径信息;
    -用户在各个关键词之间的转移次数信息。
  10. 根据权利要求1至9中任一项所述的方法,其中,所述多个对象之间的转移情况信息包括以下至少一项:
    -用户在所述多个对象中的转移路径信息;
    -用户在各个对象之间的转移次数信息;
    -用户在各个对象之间的转移概率信息。
  11. 一种用于在计算机设备中为对象确定标杆对象的装置,其中,该装置包括:
    用于获取多个对象的转移情况信息的装置,所述转移情况信息用于指示基于对象信息获取行为的、用户在所述多个对象中的转移情况;
    用于根据所述转移情况信息,为所述多个对象中的至少一个对象确定其标杆对象的装置,其中,所述标杆对象属于所述多个对象。
  12. 根据权利要求11所述的装置,其中,用于确定所述标杆对象的装置包括:
    用于根据所述转移情况信息,确定所述多个对象的排序的装置;
    用于根据所述排序,为所述多个对象中的至少一个对象确定其标杆对象的装置。
  13. 根据权利要求11或12所述的装置,其中,用于确定所述多个对象的排序的装置包括:
    用于根据所述转移情况信息以及所述多个对象的初始排序,确定所述多个对象的排序的装置。
  14. 根据权利要求13所述的装置,其中,用于确定所述多个对象的排序的装置包括:
    用于触发以下装置重复执行操作,直至满足预定条件的装置:
    用于根据所述转移情况信息以及上一次确定的所述多个对象的排序,确定所述多个对象的本次的排序的装置;
    其中,最后一次被确定的排序为所述多个对象的排序;
    其中,当确定所述多个对象的本次的排序的步骤被首次执行时,所述上一次确定的所述多个对象的排序为所述初始排序。
  15. 根据权利要求14所述的装置,其中,所述预定条件包括以下至少一项:
    -重复次数达到预定重复阈值;
    -在多次重复所述确定所述多个对象的本次的排序的步骤后,所述多个对象的排序没有发生变化。
  16. 根据权利要求11至15中任一项所述的装置,其中,所述多个对象属于同一类别。
  17. 根据权利要求11至16中任一项所述的装置,其中,用于获取所述转移情况信息的装置包括:
    用于获取多个关键词的转移情况信息的装置,其中,所述多个关键 词的转移情况信息用于指示基于对象信息获取行为的、用户在多个关键词中的转移情况;
    用于根据所述多个关键词分别关联至的对象以及所述多个关键词的转移情况信息,确定所述多个对象的转移情况信息的装置。
  18. 根据权利要求17所述的装置,其中,获取所述关键词之间的转移情况信息的装置包括:
    用于获取至少一个用户的关键词关注记录的装置,该关键词关注记录包括所述多个用户在对象信息获取行为中关注过的关键词以及所述关键词被关注的时间信息;
    用于根据所述关键词关注记录,确定所述多个关键词的转移情况信息的装置。
  19. 根据权利要求17或18所述的装置,其中,所述多个关键词的转移情况信息包括以下至少一项:
    -用户在所述多个关键词中的转移路径信息;
    -用户在各个关键词之间的转移次数信息。
  20. 根据权利要求11至19中任一项所述的装置,其中,所述多个对象之间的转移情况信息包括以下至少一项:
    -用户在所述多个对象中的转移路径信息;
    -用户在各个对象之间的转移次数信息;
    -用户在各个对象之间的转移概率信息。
  21. 一种计算机可读存储介质,所述计算机可读存储介质存储有计算机代码,当所述计算机代码被执行时,如权利要求1至10中任一项所述的方法被执行。
  22. 一种计算机程序产品,当所述计算机程序产品被计算机设备执行时,如权利要求1至10中任一项所述的方法被执行。
  23. 一种计算机设备,所述计算机设备包括存储器和处理器,所述存储器中存储有计算机代码,所述处理器被配置来通过执行所述计算机代码以执行如权利要求1至10中任一项所述的方法。
PCT/CN2015/092881 2015-02-27 2015-10-27 一种用于确定标杆对象的方法和装置 Ceased WO2016134599A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510090131.8 2015-02-27
CN201510090131.8A CN104809148B (zh) 2015-02-27 2015-02-27 一种用于确定标杆对象的方法和装置

Publications (1)

Publication Number Publication Date
WO2016134599A1 true WO2016134599A1 (zh) 2016-09-01

Family

ID=53693972

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/092881 Ceased WO2016134599A1 (zh) 2015-02-27 2015-10-27 一种用于确定标杆对象的方法和装置

Country Status (2)

Country Link
CN (1) CN104809148B (zh)
WO (1) WO2016134599A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104809148B (zh) * 2015-02-27 2018-05-25 百度在线网络技术(北京)有限公司 一种用于确定标杆对象的方法和装置

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101321190A (zh) * 2008-07-04 2008-12-10 清华大学 一种异构网络中的推荐方法及推荐系统
CN101510221A (zh) * 2009-02-17 2009-08-19 北京大学 一种用于信息检索的查询语句分析方法与系统
CN104809148A (zh) * 2015-02-27 2015-07-29 百度在线网络技术(北京)有限公司 一种用于确定标杆对象的方法和装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140114750A1 (en) * 2011-06-03 2014-04-24 Jin-Woo Jung Effective keyword selection system using keyword advertisement for internet search and an effective keyword selection method thereof
JP2013058120A (ja) * 2011-09-09 2013-03-28 Sony Corp 情報処理装置、情報処理方法、及び、プログラム

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101321190A (zh) * 2008-07-04 2008-12-10 清华大学 一种异构网络中的推荐方法及推荐系统
CN101510221A (zh) * 2009-02-17 2009-08-19 北京大学 一种用于信息检索的查询语句分析方法与系统
CN104809148A (zh) * 2015-02-27 2015-07-29 百度在线网络技术(北京)有限公司 一种用于确定标杆对象的方法和装置

Also Published As

Publication number Publication date
CN104809148A (zh) 2015-07-29
CN104809148B (zh) 2018-05-25

Similar Documents

Publication Publication Date Title
US9324112B2 (en) Ranking authors in social media systems
US9699042B2 (en) Systems and methods of classifying sessions
US8589432B2 (en) Real time searching and reporting
US9805022B2 (en) Generation of topic-based language models for an app search engine
US8965893B2 (en) System and method for grouping multiple streams of data
CN105516821B (zh) 弹幕筛选的方法及装置
US20130198240A1 (en) Social Network Analysis
CN114090850B (zh) 日志分类方法、电子设备及计算机可读存储介质
CN104092567B (zh) 确定用户的影响力排序的方法与装置
CN107153656B (zh) 一种信息搜索方法和装置
WO2018205845A1 (zh) 一种数据处理方法及服务器、计算机存储介质
JP2023534696A (ja) ネットワークトポロジーにおけるアノマリー検知
EP3061019A1 (en) Systems and methods for identifying influencers and their communities in a social data network
CN101477527A (zh) 一种检索多媒体资源的方法及装置
US10135723B2 (en) System and method for supervised network clustering
CN107077501A (zh) 将搜索结果分面
CN106959976A (zh) 一种搜索处理方法以及装置
CN112204543A (zh) 用于分布式系统的频繁模式分析
CN103744897A (zh) 故障信息的关联搜索方法、系统和网络管理系统
CN109885656B (zh) 基于量化热度的微博转发预测方法及装置
Lin et al. Predicting trends in social networks via dynamic activeness model
CN105095228A (zh) 一种社交消息的监测方法及装置
CN114491232B (zh) 信息查询方法、装置、电子设备和存储介质
US11899553B2 (en) Relevance ranking system for events identified by performance management systems
CN107077485A (zh) 用于处理图的系统和方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15883035

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15883035

Country of ref document: EP

Kind code of ref document: A1