Object classification method, device, computer readable medium and electronic device

Abstract

The embodiments of the present invention provide an object classification method, an object classification device, a computer-readable medium and an electronic device, and relate to the field of computer technology. The method comprises: determining a category tag for accessing an object to be classified according to a random walk process; determining a target category tag corresponding to the object to be classified from the category tags; and setting the target category corresponding to the target category tag as the category to which the object to be classified belongs. The technical solution of the embodiments of the present invention can overcome the problem of low object classification accuracy to a certain extent, thereby improving the accuracy and efficiency of object classification.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an object classification method, an object classification device, a computer-readable medium and an electronic device. Background Art

[0002] For e-commerce platforms, it is necessary to classify products according to different hierarchical categories to facilitate users to quickly locate the required products according to category tags. For example, users can locate shower gel products according to the category tags "personal care and cleaning", "body care" and "shower gel".

[0003] With the continuous evolution of e-commerce platforms, new products continue to emerge on e-commerce platforms, while outdated products are also eliminated from the market. With the development of business, e-commerce platforms are also constantly adjusting categories, for example, merging old categories into new categories or splitting old categories into multiple new categories. At present, the commodity classification method commonly used by e-commerce platforms is to classify commodities manually. However, manual classification has subjective factors. Along with the above-mentioned category adjustment process, staff may classify commodities that should be classified under new categories under old categories, which leads to the problem of low accuracy of commodity classification.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide an object classification method, an object classification device, a computer-readable medium and an electronic device, thereby overcoming the problem of low object classification accuracy to at least a certain extent and improving object classification accuracy and object classification efficiency.

[0006] A first aspect of an embodiment of the present invention provides an object classification method, including: determining a category tag for accessing an object to be classified according to a random walk process; determining a target category tag corresponding to the object to be classified from the category tags; and setting a target category corresponding to the target category tag as the category to which the object to be classified belongs.

[0007] In an exemplary embodiment of the present invention, before determining the category label of the object to be classified according to the random walk process, the object classification method also includes: constructing a corresponding relationship according to the attribute and the category label to execute the random walk process according to the corresponding relationship; wherein the corresponding relationship is used to represent the relationship between the object and the attribute and / or the relationship between the object and the category label; the object includes classified objects and objects to be classified.

[0008] In an exemplary embodiment of the present invention, determining a category label for accessing an object to be classified according to a random walk process includes: determining a category label for accessing an object to be classified according to a random walk process between an object and an attribute of a category label corresponding to a classified object.

[0009] In an exemplary embodiment of the present invention, determining a category signature for visiting an object to be classified according to a random walk process includes: determining a category signature for visiting the object to be classified according to executing a preset number of random walk processes in a corresponding relationship.

[0010] In an exemplary embodiment of the present invention, determining a target category signature corresponding to an object to be classified from category signatures includes: determining a target number of times a tag visits the object to be classified during a preset number of random walks; wherein the preset number of times is greater than or equal to the target number of times; determining the number of visits corresponding to each category signature that visits the object to be classified; determining a target category signature corresponding to the object to be classified from the category signatures that visit the object to be classified; wherein a ratio of the number of visits corresponding to the target category signature to the target number of visits is greater than a ratio of the number of visits corresponding to other category signatures that visit the object to be classified to the target number of visits.

[0011] In an exemplary embodiment of the present invention, the object classification method also includes: if the target category label does not exist, outputting the category label for accessing the object to be classified and prompt information for prompting manual classification; based on the detected label determination operation corresponding to the prompt information, setting the category corresponding to the label determination operation as the category to which the object to be classified belongs.

[0012] In an exemplary embodiment of the present invention, after building corresponding relationships according to the attributes and category tags, the object classification method further includes: randomly deleting a preset number of relationships from the corresponding relationships.

[0013] In an exemplary embodiment of the present invention, determining a category signature of an object to be classified according to a random walk process includes: determining a category signature of an object to be classified according to a random walk process in a corresponding relationship after deleting a preset number of relationships.

[0014] In an exemplary embodiment of the present invention, the object classification method further includes: if it is detected that there are category tags to be modified in the category tags corresponding to the classified objects, modifying the category tags to be modified according to the category tags that do not need to be modified in the category tags corresponding to the classified objects.

[0015] In an exemplary embodiment of the present invention, before determining the category label of the object to be classified according to the random walk process, the object classification method further includes: deleting the historical walk record.

[0016] According to a second aspect of an embodiment of the present invention, there is provided an object classification device, comprising: a random walk unit, for determining a category tag for accessing an object to be classified according to a random walk process; a category tag determination unit, for determining a target category tag corresponding to the object to be classified from the category tags; and a category setting unit, for setting a target category corresponding to the target category tag as the category to which the object to be classified belongs.

[0017] In an exemplary embodiment of the present invention, the object classification device further includes:

[0018] A correspondence building unit is used to build a correspondence based on the attribute and the category label before the random walk unit determines to visit the category label of the object to be classified according to the random walk process, so as to execute the random walk process according to the correspondence; wherein the correspondence is used to represent the relationship between the object and the attribute and / or the relationship between the object and the category label; the object includes the classified object and the object to be classified.

[0019] In an exemplary embodiment of the present invention, the random walk unit determines the category label of the object to be classified according to the random walk process as follows: the random walk unit determines the category label of the object to be classified according to the random walk process between the object and the attribute of the category label corresponding to the classified object.

[0020] In an exemplary embodiment of the present invention, the random walk unit determines the category signature of the object to be classified according to the random walk process as follows: the random walk unit determines the category signature of the object to be classified according to executing the random walk process a preset number of times in the corresponding relationship.

[0021] In an exemplary embodiment of the present invention, the category signature determination unit determines the target category signature corresponding to the object to be classified from the category signatures in the following manner: the category signature determination unit determines the target number of times the label visits the object to be classified during a preset number of random walks; wherein the preset number of times is greater than or equal to the target number of times; the category signature determination unit determines the number of visits corresponding to each category signature that visits the object to be classified; the category signature determination unit determines the target category signature corresponding to the object to be classified from the category signatures that visit the object to be classified; wherein the ratio of the number of visits corresponding to the target category signature to the target number of times is greater than the ratio of the number of visits corresponding to other category signatures that visit the object to be classified to the target number of times.

[0022] In an exemplary embodiment of the present invention, the object classification device further includes:

[0023] A category tag detection unit, used to detect whether a target category tag exists;

[0024] An information output unit, configured to output the category label of the object to be classified and prompt information for prompting manual classification after the category label detection unit detects that there is no target category label;

[0025] The category setting unit is further used to set the category corresponding to the label determination operation as the category to which the object to be classified belongs according to the detected label determination operation corresponding to the prompt information.

[0026] In an exemplary embodiment of the present invention, the object classification device further includes:

[0027] The random deletion unit is used to randomly delete a preset number of relationships from the corresponding relationship after the corresponding relationship construction unit constructs the corresponding relationship according to the attributes and category labels.

[0028] In an exemplary embodiment of the present invention, the random walk unit determines the category signature of the object to be classified according to the random walk process as follows: the random walk unit performs the random walk process in the corresponding relationship after deleting a preset number of relationships to determine the category signature of the object to be classified.

[0029] In an exemplary embodiment of the present invention, the category tag detection unit is further used to detect whether there is a category tag to be modified in the category tags corresponding to the classified objects; the object classification device also includes:

[0030] The label correction unit is used to correct the category signature to be corrected according to the category signatures that do not need to be corrected among the category signatures corresponding to the classified objects after the category signature detection unit detects that there is a category signature to be corrected among the category signatures corresponding to the classified objects.

[0031] In an exemplary embodiment of the present invention, the object classification device further includes:

[0032] The walk record deletion unit is used to delete the historical walk record before the random walk unit determines the category label of the object to be classified according to the random walk process.

[0033] According to a third aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored. When the program is executed by a processor, the object classification method described in the first aspect of the above embodiment is implemented.

[0034] According to a fourth aspect of an embodiment of the present invention, there is provided an electronic device, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the object classification method as described in the first aspect of the above embodiment.

[0035] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0036] In the technical solutions provided by some embodiments of the present invention, during the random walk process, the category tags (such as mobile phones, shower gels) will pass through the objects to be classified (such as commodities), and the target category tags (such as shower gels) corresponding to the objects to be classified can be determined based on the category tags of the objects to be classified. Then, the target category (such as shower gel category) corresponding to the target category tags can be set as the category to which the objects to be classified belong, which can also be understood as classifying the commodities under the shower gel category. According to the above-mentioned scheme description, on the one hand, the present invention can overcome the problem of low object classification accuracy to a certain extent, thereby improving the object classification accuracy and object classification efficiency. On the other hand, compared with the manual classification method, classifying the objects to be classified through the random walk process can reduce the classification cost.

[0037] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The drawings herein are incorporated into and constitute a part of the specification, showing embodiments consistent with the present invention, and together with the specification, are used to explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0039] Figure 1 A schematic diagram of a process flow of an object classification method according to an embodiment of the present invention is shown;

[0040] Figure 2 A schematic diagram of a corresponding relationship in an object classification method according to an embodiment of the present invention is shown;

[0041] Figure 3 A schematic diagram shows a structural block diagram of an object classification device according to an embodiment of the present invention;

[0042] Figure 4 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0043] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the present invention will be more comprehensive and complete and fully convey the concept of the example embodiments to those skilled in the art.

[0044] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present invention. However, those skilled in the art will appreciate that the technical solution of the present invention can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring various aspects of the present invention.

[0045] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0046] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0047] See also Figure 1 , Figure 1 A flow chart of an object classification method according to an embodiment of the present invention is schematically shown. The object classification method can be implemented by a server or a terminal device.

[0048] like Figure 1 As shown, the object classification method according to an embodiment of the present invention includes the following steps S110, S120 and S130, wherein:

[0049] Step S110: Determine the category label of the object to be classified according to the random walk process.

[0050] Step S120: Determine a target category tag corresponding to the object to be classified from the category tags.

[0051] Step S130: setting the target category corresponding to the target category tag as the category to which the object to be classified belongs.

[0052] The following is a detailed description of each step:

[0053] In step S110, the object to be classified can be understood as the commodity to be classified, which can be a historical commodity that has been classified or a new commodity that has not been classified. Classifying the object to be classified can be understood as reclassifying the historical commodity that has been classified but may have errors in its classification, or classifying the new commodity that has not been classified for the first time. There are one or more category tags for accessing the object to be classified, and the category tags can be pre-set tags, such as shower gel and mobile phone. In addition, the above-mentioned random walk process is that the terminal device or server uses the random walk algorithm to control the program to move randomly in the pre-built relationship network to carry the category tags to access the commodity to be classified; wherein, random walk (random walk) is also called random walk, which means that based on past performance, it is impossible to predict the future development steps and directions. Its core concept is that the conservation quantity carried by any irregular walker corresponds to a diffusion transport law, which is close to Brownian motion and is the ideal mathematical state of Brownian motion.

[0054] In step S120, if there are multiple category tags for accessing the object to be classified, the terminal device or server can determine the target category tag according to the number of accesses of each category tag, wherein the target category tag has the highest number of accesses; in addition, the category tag corresponding to the highest ratio can be determined as the target category tag according to the ratio of the number of accesses of each category tag to the total number of accesses, which is not limited in the embodiment of the present invention. If there is one category tag for accessing the object to be classified, the terminal device or server can determine whether the number of accesses of the category tag is higher than the preset number of accesses. If it is higher, it is determined as the target category tag. If it is not higher, the random walk process is continued, and when a category tag with a number of accesses higher than the preset number of accesses is detected, it is determined as the target category tag.

[0055] In step S130, the target category corresponding to the target category tag is set as the category to which the object to be classified belongs, which can also be understood as classifying the object to be classified under the target category.

[0056] It can be seen that the implementation Figure 1 The object classification method shown can overcome the problem of low object classification accuracy to a certain extent, thereby improving the object classification accuracy and object classification efficiency; in addition, compared with the traditional manual classification method, classifying the objects to be classified through a random walk process can reduce the classification cost.

[0057] As an optional implementation manner, before determining the category label of the object to be classified according to the random walk process, the object classification method further includes the following steps:

[0058] A correspondence is constructed according to the attributes and the category labels to perform a random walk process according to the correspondence; wherein the correspondence is used to represent the relationship between the object and the attribute and / or the relationship between the object and the category label; the object includes a classified object and an object to be classified.

[0059] In the embodiment of the present invention, the attribute can be understood as the attribute of the commodity, such as thickness, origin, weight and capacity. In the constructed correspondence (which can also be understood as a relationship network), there is a connection relationship between the classified object (such as a certain brand of mobile phone) and its corresponding category label (such as mobile phone), and there is also a connection relationship between the classified object (such as a certain brand of mobile phone) and its corresponding attribute (such as thickness, origin and weight), and there is a connection relationship between the unclassified object and its corresponding attribute (such as origin, weight and fragrance). In addition, it should be noted that all the connection relationships in the correspondence are non-directional.

[0060] In the embodiment of the present invention, optionally, the manner of constructing the corresponding relationship based on the attribute and the category label may specifically be: constructing a connection relationship between nodes having a corresponding relationship through the relationship between the object and the attribute and the relationship between the object and the category label, wherein the node may be an object, an attribute or a category label.

[0061] It can be seen that the implementation of this optional implementation method can facilitate the terminal device or server to perform a random walk process in the corresponding relationship by constructing a corresponding relationship, thereby achieving classification of the objects to be classified and improving the efficiency of object classification.

[0062] As another optional implementation, determining the category label of the object to be classified according to the random walk process may include the following steps:

[0063] According to the random walk process between the object and the attribute of the category label corresponding to the classified object, the category label of the object to be classified is determined.

[0064] In an embodiment of the present invention, optionally, according to the random walk process of the category tag corresponding to the classified object between the object and the attribute, the method of determining the category tag of the object to be classified can be specifically: the terminal device or server control program carries the category tag corresponding to the classified object to perform a random walk process between the object and the attribute to determine the category tag of the object to be classified. For example, the connection relationship between the object and the attribute is equivalent to a "road", and the program carrying the category tag can go to the node where the object to be classified is located through these "roads", so that the terminal device or server determines the category tag of the object to be classified.

[0065] It can be seen that by implementing this optional implementation, a random walk process can be performed in the corresponding relationship through the terminal device or the server, thereby achieving classification of the objects to be classified, thereby improving the efficiency of object classification.

[0066] As another optional implementation, determining the category label of the object to be classified according to the random walk process may include the following steps:

[0067] According to the random walk process performed a preset number of times in the corresponding relationship, the category label of the object to be classified is determined.

[0068] In an embodiment of the present invention, before the object to be classified is classified, the number of times the random walk process needs to go through, i.e., the preset number of times, can be set. Each time in the preset number of times, the program can choose to walk to the next node or stop walking this time until the preset number of random walk processes are completed. Among them, the higher the preset number of times, the more accurate the classification of the object to be classified by the terminal device or server.

[0069] It can be seen that the implementation of this optional implementation manner can improve the classification accuracy of the objects to be classified by executing a preset number of random walk processes.

[0070] As another optional implementation, determining a target category tag corresponding to the object to be classified from the category tags may include the following steps:

[0071] Determine a target number of times a tag visits an object to be classified during a preset number of random walks; wherein the preset number is greater than or equal to the target number; determine the number of visits corresponding to each category tag that visits the object to be classified; determine a target category tag corresponding to the object to be classified from the category tags that visit the object to be classified; wherein a ratio of the number of visits corresponding to the target category tag to the target number is greater than a ratio of the number of visits corresponding to other category tags that visit the object to be classified to the target number.

[0072] For example, if the category tags that visit the object to be classified include mobile phones and shower gel, where the target number of times the category tag mobile phone visits the object to be classified is 30 times, and the target number of times the category tag shower gel visits the object to be classified is 60 times, then the ratio of the number of visits of the category tag shower gel to the preset number of times (e.g., 100 times) is greater than the ratio of the category tag mobile phone to the preset number of times, indicating that the object to be classified is more likely to belong to shower gel, that is, shower gel is the target category tag.

[0073] In an embodiment of the present invention, optionally, after determining the category tag with the largest ratio of the number of visits to the preset number of visits, it can also be determined whether the largest ratio is higher than the preset ratio. If so, the category tag is determined as the target category tag corresponding to the object to be classified.

[0074] It can be seen that by implementing this optional implementation, the target category tag corresponding to the object to be classified can be determined according to the number of visits to each category tag, thereby improving the classification accuracy of the object to be classified.

[0075] As another optional implementation, the object classification method may further include the following steps:

[0076] If the target category label does not exist, the category label of the object to be classified and the prompt information for prompting manual classification are output; according to the detected label determination operation corresponding to the prompt information, the category corresponding to the label determination operation is set as the category to which the object to be classified belongs.

[0077] In the embodiment of the present invention, if the number of visits to each category tag of the object to be classified is equal, or the ratio of the category tag with the largest ratio of the number of visits to the preset number of visits is not higher than the preset ratio, it is determined that there is no target category tag, and then a prompt message is output to enable the staff to manually classify the object to be classified. The tag determination operation corresponding to the prompt message can be a non-touch operation or a touch operation, which is not limited in the embodiment of the present invention.

[0078] It can be seen that the implementation of this optional implementation can prompt the staff to perform manual classification when there is no target category label, so as to reduce the probability of under-classification or omission of the object to be classified.

[0079] As another optional implementation, after building a corresponding relationship according to the attributes and the category labels, the object classification method may further include the following steps:

[0080] Randomly delete a preset number of relations from the corresponding relations.

[0081] In the embodiment of the present invention, there are connection relationships between nodes (such as attributes, objects or category labels) and nodes in the corresponding relationship. Randomly deleting a preset number of connection relationships can improve the classification efficiency of the objects to be classified.

[0082] As another optional implementation, determining the category label of the object to be classified according to the random walk process may include the following steps:

[0083] According to executing a random walk process in the corresponding relationship after deleting a preset number of relationships, a category label for accessing the object to be classified is determined.

[0084] As another optional implementation, the object classification method may further include the following steps:

[0085] If it is detected that there is a category tag to be modified among the category tags corresponding to the classified objects, the category tags to be modified are modified according to the category tags that do not need to be modified among the category tags corresponding to the classified objects.

[0086] In an embodiment of the present invention, optionally, before the random walk process between the object and the attribute according to the category tag corresponding to the classified object, the following steps may also be included: the terminal device or the server control program walks between the object and the category tag to carry the category tag back to the node where the classified object is located, and uses the carried category tag as the category tag corresponding to the classified object; wherein the category tag may include at least one of a first-level category tag (such as personal care cleaning), a second-level category tag (such as body care), and a third-level category tag (such as shower gel), and the first-level category tag is connected to the second-level category tag, and the second-level category tag is connected to the third-level category tag. It can be seen that the implementation of this optional implementation method can perform self-inspection on the category label (such as the third-level category label) corresponding to the object according to the multi-level category label. For example, if the classified object is a brand of shower gel, its first-level category label is personal care cleaning, the second-level category label is body care, and the third-level category label is mobile phone, then the first-level category label and the second-level category label are category labels that do not need to be corrected, and the third-level category label is the category label to be corrected. The terminal device or server can correct the third-level category label to shower gel according to the first-level category label and the second-level category label.

[0087] It can be seen that by implementing this optional implementation, the category tags corresponding to the objects can be self-checked through multi-level category tags, thereby improving the classification accuracy of the objects to be classified.

[0088] As another optional implementation, before determining the category label of the object to be classified according to the random walk process, the object classification method may further include the following steps:

[0089] Delete the historical travel records.

[0090] It can be seen that by implementing this optional implementation, the corresponding relationship can be initialized by deleting the historical travel records, so as to reduce the impact of the category tag that has already visited the node where the object to be classified is located on the current classification due to the historical travel records.

[0091] See also Figure 2 , Figure 2 The following schematically shows the corresponding relationship in the object classification method according to an embodiment of the present invention. Figure 2As shown, the corresponding relationship includes the category-product subgraph and the product-attribute subgraph. The category-product subgraph is used to represent the connection relationship between the first-level category label, the second-level category label, the third-level category label and objects, and the product-attribute subgraph is used to represent the connection relationship between objects and object attributes. Among them, the first-level category labels include digital and personal care cleaning, the second-level category labels include mobile communication and body care, the third-level category labels include mobile phones and shower gel, the objects include object 1, object 2, object 3 and object 4, and the object attributes include optical image stabilization, thickness, memory, origin, weight, capacity and fragrance. It should be noted that Figure 2 The corresponding relationship shown in the figure is based on the category label and object attribute shown in the figure as an example, and the embodiment of the present invention is not limited to application in Figure 2 In the corresponding relationship shown.

[0092] like Figure 2 As shown, the third-level category labels corresponding to objects 1, 2, and 3 are mobile phone, mobile phone, and shower gel, respectively. Object 4 has no corresponding third-level category label. The object classification method disclosed in the embodiment of the present invention can classify object 4. Specifically, the terminal device or server can control the program to Figure 2 A random walk process is performed in the corresponding relationship shown. For example, when the program carries a category tag mobile phone at the node where object 1 is located, it can choose to walk to the node connected to object 1 (such as optical image stabilization, thickness, memory or origin) or choose to stop at the object 1 node; after experiencing a preset number of random walk processes, the target category tag corresponding to object 4 (such as shower gel) can be determined according to the number of visits to each tag of object 4, and object 4 can be classified into the target category corresponding to the target category tag.

[0093] visible, Figure 2 The corresponding relationship diagram shown can overcome the problem of low object classification accuracy to a certain extent, thereby improving the object classification accuracy and object classification efficiency; in addition, compared with the traditional manual classification method, classifying the objects to be classified through a random walk process can reduce the classification cost.

[0094] See also Figure 3 , Figure 3 The structure block diagram of the object classification device according to the embodiment of the present invention is schematically shown. The object classification device comprises: a random walk unit 301 , a category label determination unit 302 and a category setting unit 303 .

[0095] Among them, the random walk unit 301 is used to determine the category tag of the object to be classified according to the random walk process; the category tag determination unit 302 is used to determine the target category tag corresponding to the object to be classified from the category tags; the category setting unit 303 is used to set the target category corresponding to the target category tag as the category to which the object to be classified belongs.

[0096] It can be seen that the implementation Figure 3 The object classification device shown can overcome the problem of low object classification accuracy to a certain extent, thereby improving the object classification accuracy and object classification efficiency; in addition, compared with the traditional manual classification method, classifying the objects to be classified through a random walk process can reduce the classification cost.

[0097] As an optional implementation, the object classification device may further include a corresponding relationship building unit (not shown), wherein:

[0098] The correspondence construction unit is used to construct a correspondence according to the attribute and the category label before the random walk unit 301 determines to visit the category label of the object to be classified according to the random walk process, so as to execute the random walk process according to the correspondence; wherein the correspondence is used to represent the relationship between the object and the attribute and / or the relationship between the object and the category label; the object includes the classified object and the object to be classified.

[0099] It can be seen that the implementation of this optional implementation method can facilitate the terminal device or server to perform a random walk process in the corresponding relationship by constructing a corresponding relationship, thereby achieving classification of the objects to be classified and improving the efficiency of object classification.

[0100] As another optional implementation, the random walk unit 301 determines the category label of the object to be classified according to the random walk process. Specifically, the random walk unit 301 determines the category label of the object to be classified according to the random walk process between the object and the attribute of the category label corresponding to the classified object.

[0101] It can be seen that by implementing this optional implementation, a random walk process can be performed in the corresponding relationship through the terminal device or the server, thereby achieving classification of the objects to be classified, thereby improving the efficiency of object classification.

[0102] As another optional implementation, the random walk unit 301 determines the category signature of the object to be classified according to the random walk process. Specifically, the random walk unit 301 determines the category signature of the object to be classified by performing a preset number of random walk processes in the corresponding relationship.

[0103] It can be seen that the implementation of this optional implementation manner can improve the classification accuracy of the objects to be classified by executing a preset number of random walk processes.

[0104] As another optional implementation, the category signature determination unit 302 determines the target category signature corresponding to the object to be classified from the category signatures in a specific manner as follows: the category signature determination unit 302 determines the target number of times the tag visits the object to be classified during a preset number of random walks; wherein the preset number of times is greater than or equal to the target number of times; the category signature determination unit 302 determines the number of visits corresponding to each category signature that visits the object to be classified; the category signature determination unit 302 determines the target category signature corresponding to the object to be classified from the category signatures that visit the object to be classified; wherein the ratio of the number of visits corresponding to the target category signature to the target number of times is greater than the ratio of the number of visits corresponding to other category signatures that visit the object to be classified to the target number of times.

[0105] It can be seen that by implementing this optional implementation, the target category tag corresponding to the object to be classified can be determined according to the number of visits to each category tag, thereby improving the classification accuracy of the object to be classified.

[0106] As another optional implementation, the object classification device may further include a category tag detection unit (not shown) and an information output unit (not shown), wherein:

[0107] A category tag detection unit, used to detect whether a target category tag exists;

[0108] An information output unit, configured to output the category label of the object to be classified and prompt information for prompting manual classification after the category label detection unit detects that there is no target category label;

[0109] The category setting unit 303 is further configured to set the category corresponding to the label determination operation as the category to which the object to be classified belongs according to the detected label determination operation corresponding to the prompt information.

[0110] It can be seen that the implementation of this optional implementation can prompt the staff to perform manual classification when there is no target category label, so as to reduce the probability of under-classification or omission of the object to be classified.

[0111] As another optional implementation, the object classification device may further include a random deletion unit (not shown), wherein:

[0112] The random deletion unit is used to randomly delete a preset number of relationships from the corresponding relationship after the corresponding relationship construction unit constructs the corresponding relationship according to the attributes and category labels.

[0113] It can be seen that implementing this optional implementation can improve the classification efficiency of the objects to be classified.

[0114] As another optional implementation, the random walk unit 301 determines the category signature of the object to be classified according to the random walk process by executing the random walk process in the corresponding relationship after deleting a preset number of relationships to determine the category signature of the object to be classified.

[0115] As another optional implementation, the object classification device may further include a label correction unit (not shown), wherein:

[0116] The category tag detection unit is further used to detect whether there is a category tag to be corrected in the category tags corresponding to the classified objects;

[0117] The label correction unit is used to correct the category signature to be corrected according to the category signatures that do not need to be corrected among the category signatures corresponding to the classified objects after the category signature detection unit detects that there is a category signature to be corrected among the category signatures corresponding to the classified objects.

[0118] It can be seen that by implementing this optional implementation, the category tags corresponding to the objects can be self-checked through multi-level category tags, thereby improving the classification accuracy of the objects to be classified.

[0119] As another optional implementation, the object classification device may further include a wandering record deletion unit (not shown), wherein:

[0120] The walking record deleting unit is used to delete the historical walking record before the random walking unit 301 determines the category label of the object to be classified according to the random walking process.

[0121] It can be seen that by implementing this optional implementation, the corresponding relationship can be initialized by deleting the historical travel records, so as to reduce the impact of the category tag that has already visited the node where the object to be classified is located on the current classification due to the historical travel records.

[0122] Since the various functional modules of the data transmission device of the exemplary embodiment of the present invention correspond to the steps of the exemplary embodiment of the control method based on virtual operations in the game, for details not disclosed in the embodiment of the device of the present invention, please refer to the embodiment of the control method of virtual operations in the game of the present invention mentioned above.

[0123] See also Figure 4 , which shows a schematic diagram of the structure of a computer system 400 of an electronic device suitable for implementing an embodiment of the present invention. Figure 4 The computer system 400 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0124] like Figure 4 As shown, computer system 400 includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 402 or a program loaded from a storage part 408 into a random access memory (RAM) 403. Various programs and data required for system operation are also stored in RAM 403. CPU 401, ROM 402 and RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to bus 404.

[0125] The following components are connected to the I / O interface 405: an input section 406 including a keyboard, a mouse, etc.; an output section 407 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed, so that a computer program read therefrom is installed into the storage section 408 as needed.

[0126] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 409, and / or installed from a removable medium 411. When the computer program is executed by a central processing unit (CPU) 401, the above-mentioned functions defined in the system of the present application are executed.

[0127] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0128] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0129] The units involved in the embodiments of the present invention may be implemented by software or hardware, and the units described may also be arranged in a processor. The names of these units do not, in some cases, limit the units themselves.

[0130] As another aspect, the present application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the object classification method described in the above embodiment.

[0131] For example, the electronic device can implement Figure 1 As shown in: step S110, determining the category tag of the object to be classified according to the random walk process; step S120, determining the target category tag corresponding to the object to be classified from the category tags; step S130, setting the target category corresponding to the target category tag as the category to which the object to be classified belongs.

[0132] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0133] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the implementation of the present invention.

[0134] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed by the present invention. The specification and examples are to be considered exemplary only, and the true scope and spirit of the present invention are indicated by the following claims.

[0135] It should be understood that the present invention is not limited to the exact construction that has been described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. An object classification method, characterized in that: include: Determine the category label of the object to be classified according to the random walk process; Determine a target category tag corresponding to the object to be classified from the category tags; Setting the target category corresponding to the target category tag as the category to which the object to be classified belongs; wherein the object to be classified is a commodity to be classified; The step of determining the category label of the object to be classified according to the random walk process includes: Determine the category label of the object to be classified according to the random walk process between the object and the attribute of the category label corresponding to the classified object; wherein the attribute includes the attribute of the commodity, and the attribute of the commodity includes at least one of the following: origin, weight, and capacity; Before the random walk process between objects and attributes according to the category labels corresponding to the classified objects, it also includes: The control program moves between the object and the category tag to carry the category tag back to the node where the classified object is located, and uses the carried category tag as the category tag corresponding to the classified object; wherein the category tag includes at least one of a primary category tag, a secondary category tag, and a tertiary category tag, the primary category tag is connected to the secondary category tag, and the secondary category tag is connected to the tertiary category tag; If it is detected that there is a category tag to be modified in the category tags corresponding to the classified objects, the category tags to be modified are modified according to the category tags that do not need to be modified in the category tags corresponding to the classified objects.

2. The method according to claim 1, characterized in that: Before determining the category label of the object to be classified according to the random walk process, the method further includes: A correspondence is constructed according to the attribute and the category label to perform a random walk process according to the correspondence; wherein the correspondence is used to represent the relationship between the object and the attribute and / or the relationship between the object and the category label; the object includes a classified object and an object to be classified.

3. The method according to claim 2, characterized in that The step of determining the category label of the object to be classified according to the random walk process includes: According to the random walk process executed a preset number of times in the corresponding relationship, a category label of the object to be classified is determined.

4. The method according to claim 3, characterized in that The step of determining a target category tag corresponding to the object to be classified from the category tags includes: Determine a target number of times that a tag visits the object to be classified during the preset number of random walks; wherein the preset number of times is greater than or equal to the target number of times; Determine the number of visits corresponding to each category tag of the object to be classified; A target category tag corresponding to the object to be classified is determined from the category tags of the objects to be classified; wherein the ratio of the number of visits corresponding to the target category tag to the target number of visits is greater than the ratio of the number of visits corresponding to other category tags of the objects to be classified to the target number of visits.

5. The method according to claim 4, characterized in that The method further comprises: If the target category label does not exist, outputting the category label of the object to be classified and prompting manual classification; According to the detected label determination operation corresponding to the prompt information, the category corresponding to the label determination operation is set as the category to which the object to be classified belongs.

6. The method according to claim 2, characterized in that After the corresponding relationship is constructed according to the attribute and the category label, the method further includes: A preset number of relationships are randomly deleted from the corresponding relationships.

7. The method according to claim 6, characterized in that The step of determining the category label of the object to be classified according to the random walk process includes: According to executing a random walk process in the corresponding relationship after deleting the preset number of relationships, a category label for accessing the object to be classified is determined.

8. The method according to any one of claims 2 to 7, characterized in that: Before determining the category label of the object to be classified according to the random walk process, the method further includes: Delete the historical travel records.

9. An object classification device, characterized in that: include: A random walk unit, used for determining a category label of an object to be classified according to a random walk process; A category label determination unit, configured to determine a target category label corresponding to the object to be classified from the category labels; A category setting unit, used to set the target category corresponding to the target category tag as the category to which the object to be classified belongs; wherein the object to be classified is a commodity to be classified; The random walk unit determines the category label of the object to be classified according to the random walk process, including: Determine the category label of the object to be classified according to the random walk process between the object and the attribute of the category label corresponding to the classified object; wherein the attribute includes the attribute of the commodity, and the attribute of the commodity includes at least one of the following: origin, weight, and capacity; Before the random walk process between the object and the attribute according to the category label corresponding to the classified object, the device is also used for: The control program moves between the object and the category tag to carry the category tag back to the node where the classified object is located, and uses the carried category tag as the category tag corresponding to the classified object; wherein the category tag includes at least one of a primary category tag, a secondary category tag, and a tertiary category tag, the primary category tag is connected to the secondary category tag, and the secondary category tag is connected to the tertiary category tag; If it is detected that there is a category tag to be modified in the category tags corresponding to the classified objects, the category tags to be modified are modified according to the category tags that do not need to be modified in the category tags corresponding to the classified objects.

10. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the object classification method according to any one of claims 1 to 8 is implemented.

11. An electronic device, characterized in that: include: one or more processors; A storage device, used for storing one or more programs, when the one or more programs are executed by the one or more processors, enables the one or more processors to implement the object classification method according to any one of claims 1 to 8.

Images