An item category normalization method and device applied to cross-platforms

Abstract

The application discloses a kind of applied to cross-platform article category normalization method and device, it is related to computer technology field.The specific embodiment of the method includes: the information related to first article and second article is collected respectively, whether the first article and the second article belong to standard product is inquired;If both belong to, then using hard matching algorithm, the information of the first article and second article is carried out category normalization processing;If both do not belong to, then using similarity matching algorithm, the information of the first article and second article is carried out category normalization processing;Output the processing result that the first article and the second article belong to same category or different category.The embodiment belongs to the same SKU cross-platform article and is classified as a class, so as to realize the standardization of different platform article category management.

Description

TECHNICAL FIELD

[0001] The present application relates to the computer technical field, and particularly relates to a kind of applied to cross-platform article category normalization method and device. BACKGROUND

[0002] With the extensive application of Internet, people's production and life style are also undergoing profound changes, and the attention to platform, especially application platform, is increasing. However, there are numerous application platforms in the market at present, and the article categories set by each website are in chaos. Considering that article category normalization is of great significance in website management, such as quick searching of target category by supply / demand parties, defining the service range of website, etc., the article categories in each platform need to be normalized and managed. SUMMARY

[0003] Therefore, the embodiments of the present application provide a kind of applied to cross-platform article category normalization method and device, which can solve the chaotic phenomenon of article category management of each platform in the prior art.

[0004] To achieve the above object, according to one aspect of the embodiments of the present application, a kind of applied to cross-platform article category normalization method is provided, comprising:

[0005] Collecting information related to a first article and a second article respectively, and querying whether the first article and the second article belong to a standard article; wherein the first article and the second article belong to different platforms;

[0006] If both belong to, a hard matching algorithm is used to normalize the information of the first article and the second article; or

[0007] If both do not belong to, a similarity matching algorithm is used to normalize the information of the first article and the second article;

[0008] Output the processing result that the first article and the second article belong to the same category or different categories.

[0009] Optionally, the article information includes brand and specification parameter, and the specification parameter includes model;

[0010] The hard matching algorithm is used to normalize the information of the first article and the second article, comprising:

[0011] Judging whether the model and the brand of the first article and the second article are the same, if both are the same, it is determined that the first article and the second article belong to the same category, otherwise belong to different categories.

[0012] Optionally, the first article and the second article are electronic products, and the specification parameter further includes memory model ROM value.

[0013] The judging whether the model and the brand of the first item and the second item are same comprises:

[0014] The judging whether the model, the ROM value and the brand of the first item and the second item are same.

[0015] Optionally, the item information comprises a title.

[0016] Before the judging whether the model and the brand of the first item and the second item are same, the method further comprises:

[0017] For any one of the first item and the second item, if there is no model in the rule parameter, the model is obtained from the title of the any one by a cutting method; and

[0018] If the model formats of the first item and the second item are different, the model of the first item and the second item is normalized according to a preset model format.

[0019] Optionally, the adopting a similarity matching algorithm to perform category normalization processing on the information of the first item and the second item comprises:

[0020] Calculating the similarity of the title of the first item and the title of the second item, if the similarity is less than a first preset similarity threshold, it is determined that the first item and the second item belong to different categories;

[0021] Otherwise, calculating the similarity of the model of the first item and the model of the second item, if the similarity is less than a second preset similarity threshold, it is determined that the first item and the second item belong to different categories, otherwise, they belong to the same category.

[0022] Optionally, the calculating the similarity of the title of the first item and the title of the second item comprises:

[0023] Extracting a first feature from the title of the first item, and generating a first text feature vector based on the first feature; and

[0024] Extracting a second feature from the title of the second item, and generating a second text feature vector based on the second feature;

[0025] Calculating the similarity of the first text feature vector and the second text feature vector.

[0026] Optionally, before the calculating the similarity of the title of the first item and the title of the second item, the method further comprises:

[0027] The title of any of the first and second items is cut by using a title cutting method;

[0028] If the brand name does not exist in the cut word, the brand name to which the any item belongs is added to the title of the any item.

[0029] Optionally, if the similarity is less than the second preset similarity threshold, it is determined that the first item and the second item belong to different categories, and the method further comprises:

[0030] If the similarity is less than the second preset similarity threshold, the difference between the value of the first item and the value of the second item is calculated.

[0031] If the difference is within a preset difference range, it is determined that the first item and the second item belong to the same category, otherwise they belong to different categories.

[0032] To achieve the above object, according to another aspect of the embodiment of the present application, a kind of device for cross-platform item category is provided, comprising:

[0033] The query module is used to collect information related to the first item and the second item respectively, and query whether the first item and the second item belong to the same category; wherein the first item and the second item belong to different platforms;

[0034] The hard matching module is used to determine whether the first item and the second item belong to the same category by using a hard matching algorithm if they belong to the same category; or

[0035] The similarity matching module is used to determine whether the first item and the second item belong to the same category by using a similarity matching algorithm if they do not belong to the same category.

[0036] The output module is used to output the processing result of whether the first item and the second item belong to the same category or different categories.

[0037] Optionally, the item information includes brand and specification parameters, and the specification parameters include model number.

[0038] The hard matching module is used to determine whether the model number and the brand of the first item and the second item are the same, and if they are the same, it is determined that the first item and the second item belong to the same category, otherwise they belong to different categories.

[0039] Optionally, the first item and the second item are electronic products, and the specification parameters further include ROM value.

[0040] The hard matching module is configured to determine whether the model, ROM value, and brand of the first item and the second item are the same.

[0041] Optionally, the item information includes a title.

[0042] The hard matching module is further configured to:

[0043] For any one of the first item and the second item, if there is no model in the rule parameter, the model is obtained from the title of the any one item by word segmentation.

[0044] If the model formats of the first item and the second item are different, the models of the first item and the second item are normalized according to a preset model format.

[0045] Optionally, the similarity matching module is configured to:

[0046] Calculate the similarity of the title of the first item and the title of the second item, and if the similarity is less than a first preset similarity threshold, determine that the first item and the second item belong to different categories.

[0047] Otherwise, calculate the similarity of the model of the first item and the model of the second item, and if the similarity is less than a second preset similarity threshold, determine that the first item and the second item belong to different categories, otherwise belong to the same category.

[0048] Optionally, the similarity matching module is configured to:

[0049] Extract a first feature from the title of the first item, and generate a first text feature vector based on the first feature; and

[0050] Extract a second feature from the title of the second item, and generate a second text feature vector based on the second feature.

[0051] Calculate the similarity of the first text feature vector and the second text feature vector.

[0052] Optionally, the similarity matching module is further configured to:

[0053] For any one of the first item and the second item, perform word segmentation processing on the title of the any one item by using a title word segmentation comparison device.

[0054] If the brand name does not exist in the segmented words obtained by word segmentation, add the brand name to which the any one item belongs to the title of the any one item.

[0055] Optionally, the similarity matching module is further configured to:

[0056] If the similarity is less than a second preset similarity threshold, a difference between the value of the first item and the value of the second item is calculated;

[0057] If the difference is within a preset difference range, the first item and the second item are determined to belong to the same category, otherwise, the first item and the second item belong to different categories.

[0058] To achieve the above object, according to another aspect of the embodiment of the present application, an electronic device for item category unification is provided.

[0059] The electronic device of the embodiment of the present application comprises one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned any one of the application of the cross-platform item category unification method.

[0060] To achieve the above object, according to another aspect of the embodiment of the present application, a computer readable medium having a computer program stored thereon is provided, the program is executed by a processor to implement the above-mentioned any one of the application of the cross-platform item category unification method.

[0061] According to the scheme provided by the present application, one embodiment of the above-mentioned application has the following advantages or beneficial effects: the item information collected by multiple platforms is classified into the same category according to whether it belongs to a standard item, and a hard matching algorithm or a similarity matching algorithm is used to achieve the purpose of cross-platform category normalization, thereby realizing the standardization of item category management of different platforms.

[0062] The further effects of the above-mentioned non-conventional optional mode will be described in the following combined with the specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0063] The accompanying drawings are used to better understand the present application, and do not constitute an improper limitation on the present application. Among them:

[0064] Figure 1 is a main flowchart of an application of a cross-platform item category unification method according to an embodiment of the present application;

[0065] Figure 2 is a flowchart of an optional application of a cross-platform item category unification method according to an embodiment of the present application;

[0066] Figure 3 is a flowchart of another optional application of a cross-platform item category unification method according to an embodiment of the present application;

[0067] Figure 4is a flowchart of a specific application of the item category normalization method across platforms according to an embodiment of the present application;

[0068] Figure 5 is a schematic diagram of the main modules of an item category normalization device across platforms according to an embodiment of the present application;

[0069] Figure 6 is an exemplary system architecture diagram to which embodiments of the present application can be applied;

[0070] Figure 7 is a structural diagram of a computer system of a mobile device or server suitable for implementing embodiments of the present application. DETAILED DESCRIPTION

[0071] Exemplary embodiments of the present application are described herein with reference to the accompanying drawings, which are meant to be exemplary and not limiting. Therefore, it should be recognized that many changes and modifications can be made to the embodiments described herein, without departing from the scope and spirit of the application. Also, for the sake of brevity and clarity, descriptions of well-known functions and constructions are omitted herein.

[0072] Brand: Each brand represents different product characteristics / cultural background / design concept / psychological target. Establishing a brand can be used to identify a certain product or service for sale. Not only to sell products to users, but also to make users like the product through use, constantly promote, form brand loyalty, and then promote repeated purchases. Some enterprises have set up a good image for their brands, given a beautiful emotion, or represented a certain culture, so the brand adds value to the product, and the brand and the brand product form a beautiful memory in the user's mind.

[0073] Model: It is a specific supplement and explanation of the brand, and a more powerful explanation of the influence of a product brand.

[0074] Referring to Figure 1 , a main flowchart of an item category normalization method across platforms is shown, which includes the following steps:

[0075] S101: Collect information related to a first item and a second item respectively, and query whether the first item and the second item belong to a standard item; wherein the first item and the second item belong to different platforms;

[0076] S102: If both belong to, then use a hard matching algorithm to perform category normalization processing on the information of the first item and the second item;

[0077] S103: If none, a similarity matching algorithm is used to normalize the categories of the information of the first and second items;

[0078] S104: Output the processing result that the first item and the second item belong to the same category or different categories.

[0079] In the above embodiment, for step S101, first, information related to a plurality of items in each platform is collected, and the collected information is preprocessed and cleaned through a preset dirty data (Dirty Read) library to filter out dirty data, such as

self-operated

[0080] In actual operation, considering that the amount of collected item information is large and offline, HDFS is preferred to store the item information, and other databases can be used in the case of small amount of information. The present scheme preferably uses a Hive data warehouse to extract, transform, and load data, which is a mechanism that can store, query, and analyze large-scale data stored in Hadoop.

[0081] For steps S102-S104, the items are divided into marked items and non-marked items in advance on the basis of categories, the marked items: the information of item brands, titles, specification parameters, etc. is relatively complete, standard, and unified, such as 3C digital products; and the non-marked items: relative to the marked items.

[0082] Since the item information of the marked category is relatively clear and unified, a hard matching algorithm is used for category normalization processing, that is, by comparing the brand, model, specification parameter, etc. of the item, it is directly determined whether the first item and the second item belong to the same SKU (Stock Keeping Unit, inventory unit), see the following Figure 2 description; and the non-marked item is not clear due to the confusion of related information, so a similarity matching algorithm is used for category normalization processing, that is, by calculating the text similarity of the item title and the model similarity to determine whether they belong to the same SKU, see the following Figure 3 description.

[0083] It should be noted that for the case where only one of the first item and the second item belongs to the marked item, no processing is performed, or it is directly determined that the two do not belong to the same category.

[0084] Finally, the IDs, URLs and titles of the normalized items belonging to the same category are outputted and stored in the Hive data warehouse, so that the parties involved can quickly find the interested category; and the platform can accurately locate the scope of its own services.

[0085] The method provided in the above embodiment collects item information collected by multiple platforms, and uses a corresponding algorithm to classify items belonging to the same SKU into a category according to whether they all belong to a marked item, so as to achieve the purpose of cross-platform category normalization, thereby realizing the standardization of item category management of different platforms.

[0086] Referring to Figure 2 , an optional cross-platform item category normalization method flowchart according to an embodiment of the present application is shown, which includes the following steps:

[0087] S201: Collect information related to a first item and a second item respectively, and query whether the first item and the second item both belong to a marked item; wherein the first item and the second item belong to different platforms;

[0088] S202: If both belong to, determine whether the model and the brand of the first item and the second item are the same;

[0089] S203: If both are the same, it is determined that the first item and the second item belong to the same category, otherwise they belong to different categories;

[0090] S204: If both do not belong to, a similarity matching algorithm is used to perform category normalization processing on the information of the first item and the second item;

[0091] S205: Output the processing result of whether the first item and the second item belong to the same category or different categories.

[0092] In the above embodiment, for steps S201, S204 and S205, refer to the description of steps S101, S103 and S104 shown in Figure 1 , which will not be repeated here.

[0093] In the above embodiment, for steps S202 and S203, the present embodiment is used to perform category normalization processing on marked items by a hard matching algorithm, including data preprocessing, logical judgment and normalized SKU result output:

[0094] 1) Classification processing: the items are classified by brand in advance, and multiple items under the same brand are obtained, so that the first item and the second item in the present solution belong to products under the same brand although they belong to different platforms;

[0095] Title processing: the title of the article is segmented, stop words are removed, and the segmented results are reordered according to the order of the words in the original title to generate a new title;

[0096] Model processing: the model name in the article specification parameter is normalized, such as case conversion, space, and redundant word processing. If there is no model in the rule parameter, it is obtained by cutting from the title. In addition, if the models of the same article are not uniform, they are normalized according to the preset model format, for example, honor 8, honor 8 (FRD-AL10), and FRD-AL10 are normalized to honor 8. Further, the articles can be classified according to the model to obtain multiple articles under the same model.

[0097] 2) For the first article and the second article, if the model and the brand of the two are the same, it is determined that the two belong to the same category, otherwise they belong to different categories. For electronic products, the ROM (ReadOnly Memory, Read Only Memory) value can also be obtained from the article specification parameter, at this time, it is necessary to determine whether the model, ROM value and brand of the first article and the second article are the same, if they are the same, it is determined that the two belong to the same category, otherwise they belong to different categories.

[0098] The method provided by the above embodiment, in the case that the first article and the second article belong to the standard article, the hard matching algorithm is used for category normalization processing, which provides a basis for subsequent rapid release and management of articles, and facilitates the procurement party to quickly find the needed articles / services.

[0099] Referring to Figure 3 , another optional application of the article category normalization method flowchart across platforms according to an embodiment of the application is shown, which includes the following steps:

[0100] S301: Collect information related to the first article and the second article respectively, and query whether the first article and the second article belong to the standard article; wherein the first article and the second article belong to different platforms;

[0101] S302: If they belong to the standard article, a hard matching algorithm is used to perform category normalization processing on the information of the first article and the second article;

[0102] S303: If they do not belong to the standard article, the similarity of the title of the first article and the title of the second article is calculated, and if the similarity is less than a first preset similarity threshold, it is determined that the first article and the second article belong to different categories;

[0103] S304: Otherwise, the similarity of the model of the first article and the model of the second article is calculated;

[0104] S305: If the similarity is less than a second preset similarity threshold, it is determined that the first item and the second item belong to different categories, otherwise belong to the same category;

[0105] S306: Output the processing result that the first item and the second item belong to the same category or different categories.

[0106] For steps S301, S302 and S306 in the above embodiment, refer to the description of steps S101, S102 and S104 shown in FIG. 1, which will not be repeated here. Figure 1

[0107] For steps S303-S305 in the above embodiment, the present embodiment is used for category normalization processing of non-standard items through a similarity matching algorithm, that is, whether the first item and the second item belong to the same SKU is determined by calculating whether the text similarity and the model similarity of the item title meet the requirements, as follows:

[0108] 1) Fill in the missing "BrandID" value in the item information: adopt a title word segmentation comparison method to perform word segmentation processing on the item title, and if there is no brand name, add the brand name of the item to the title. For example, after word segmentation of "2013 winter new mink collar black slim warm women's cotton coat", "2012 winter new mink collar black slim warm women's cotton coat" is obtained, which does not contain the brand name A, and then the brand name A is added to the title.

[0109] 2) Perform word segmentation processing on the title to remove interfering words, and then select features such as "url", "title", "price", "productID", "BrandID" from the title, and convert them into a text feature vector using a tf-idf (term frequency-inverted document frequency) algorithm;

[0110] 3) Calculate the similarity between the first text feature vector of the first item and the second text feature vector of the second item, and if it is greater than or equal to a first preset similarity threshold, continue to step 4), otherwise it is determined that the two items belong to different categories;

[0111] 4) Calculate the similarity of the first model of the first item and the model of the second item, and if it is greater than or equal to a second preset similarity threshold, it is determined that the two items belong to the same category, otherwise it is determined that they belong to different categories.

[0112] ​As further optimization, the scheme can also judge the similarity of the goods according to the price difference of the first goods and the second goods, if the prices of the two are the same or the difference is within a preset difference range, it is determined that the two belong to the same category, otherwise they belong to different categories. And the original price of the goods after category normalization can reach normalization, and the order payment price is the price after using the discount activity.

[0113] The method provided by the above embodiment, in the case that neither the first goods nor the second goods belong to the standard goods, the similarity matching algorithm is used for category normalization processing, which provides a basis for subsequent rapid release and management of goods, and facilitates the procurement party to quickly find the needed goods / service.

[0114] Referring to Figure 4 , a method flowchart of the goods category normalization method applied to cross-platform is shown, including the following steps:

[0115] S401: Collect information related to the first goods and the second goods respectively, and query whether the first goods and the second goods belong to the standard goods; wherein the first goods and the second goods belong to different platforms;

[0116] S402: If both belong to the standard goods, judge whether the model and the brand of the first goods and the second goods are the same;

[0117] S403: If both are the same, it is determined that the first goods and the second goods belong to the same category, otherwise they belong to different categories;

[0118] S404: If neither belongs to the standard goods, calculate the similarity of the title of the first goods and the title of the second goods, if the similarity is less than a first preset similarity threshold, it is determined that the first goods and the second goods belong to different categories;

[0119] S405: Otherwise, calculate the similarity of the model of the first goods and the model of the second goods;

[0120] S406: If the similarity is less than a second preset similarity threshold, it is determined that the first goods and the second goods belong to different categories, otherwise they belong to the same category;

[0121] S407: Output the processing result that the first goods and the second goods belong to the same category or different categories.

[0122] Referring to Figure 5 , a main module schematic diagram of the device 500 for applying the goods category normalization method to cross-platform is shown, which includes:

[0123] The query module 501 is configured to collect information related to the first and second items respectively, and query whether the first and second items belong to the same category.

[0124] The hard matching module 502 is configured to, if the first and second items belong to the same category, perform category normalization processing on the information of the first and second items by using a hard matching algorithm.

[0125] The similarity matching module 503 is configured to, if the first and second items do not belong to the same category, perform category normalization processing on the information of the first and second items by using a similarity matching algorithm.

[0126] The output module 504 is configured to output a processing result of whether the first and second items belong to the same category or different categories.

[0127] In the device, the item information includes a brand and a specification parameter, and the specification parameter includes a model number.

[0128] The hard matching module 502 is configured to determine whether the model number and the brand of the first and second items are the same, and if the model number and the brand of the first and second items are the same, determine that the first and second items belong to the same category, otherwise, belong to different categories.

[0129] In the device, the first and second items are electronic products, and the specification parameter further includes a memory model ROM value.

[0130] The hard matching module 502 is configured to determine whether the model number, the ROM value and the brand of the first and second items are the same.

[0131] In the device, the item information includes a title.

[0132] The hard matching module 502 is further configured to:

[0133] For any one of the first and second items, if the model number is not included in the specification parameter, the model number is obtained from the title of the any one item by using a word segmentation method.

[0134] If the model number formats of the first and second items are different, the model numbers of the first and second items are normalized according to a preset model number format.

[0135] In the device, the similarity matching module 503 is configured to:

[0136] Calculate the similarity of the title of the first item and the title of the second item, if the similarity is less than a first preset similarity threshold, determine that the first item and the second item belong to different categories.

[0137] Otherwise, calculate the similarity of the model of the first item and the model of the second item, if the similarity is less than a second preset similarity threshold, determine that the first item and the second item belong to different categories, otherwise belong to the same category.

[0138] In the device of the embodiment of the application, the similarity matching module 503 is configured to:

[0139] Extract a first feature from the title of the first item, and generate a first text feature vector based on the first feature; and

[0140] Extract a second feature from the title of the second item, and generate a second text feature vector based on the second feature.

[0141] Calculate the similarity of the first text feature vector and the second text feature vector.

[0142] In the device of the embodiment of the application, the similarity matching module 503 is further configured to:

[0143] For any one of the first item and the second item, perform word segmentation processing on the title of the any one item by using a title word segmentation comparison device.

[0144] If there is no brand name in the segmented words obtained by word segmentation, add the brand name to which the any one item belongs to the title of the any one item.

[0145] In the device of the embodiment of the application, the similarity matching module 503 is further configured to:

[0146] If the similarity is less than a second preset similarity threshold, calculate the difference between the value of the first item and the value of the second item.

[0147] If the difference is within a preset difference range, determine that the first item and the second item belong to the same category, otherwise belong to different categories.

[0148] In addition, the specific implementation of the device in the embodiment of the application has been described in detail in the above method, and therefore the repeated content will not be described here.

[0149] Figure 6 An exemplary system architecture 600 to which embodiments of the application can be applied is shown.

[0150] As Figure 6As shown, the system architecture 600 can include terminal devices 601, 602, 603, a network 604 and a server 605 (only examples). The network 604 is used to provide a communication link medium between the terminal devices 601, 602, 603 and the server 605. The network 604 can include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0151] The user can use the terminal devices 601, 602, 603 to interact with the server 605 through the network 604 to receive or send messages, etc. Various communication client applications can be installed on the terminal devices 601, 602, 603.

[0152] The terminal devices 601, 602, 603 can be various electronic devices with display screens and supporting web browsing, and the server 605 can be a server providing various services.

[0153] It should be noted that the method provided by the embodiments of the present application is generally executed by the server 605, and accordingly, the apparatus is generally provided in the server 605.

[0154] It should be understood that, Figure 6 The number of terminal devices, networks and servers in the above description is only illustrative. According to the implementation needs, there can be any number of terminal devices, networks and servers.

[0155] The following refers to Figure 7 which shows a structural schematic diagram of a computer system 700 of a terminal device suitable for implementing the embodiments of the present application. Figure 7 The terminal device shown is only an example and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0156] As shown in Figure 7 , the computer system 700 includes a central processing unit (CPU) 701 which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 702 or programs loaded from a storage portion 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the system 700 are also stored. The CPU 701, the ROM 702 and the RAM 703 are connected to each other through a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0157] The following components are connected to the I / O interface 705: an input part 706 including a keyboard, a mouse, etc.; an output part 707 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage part 708 including a hard disk, etc.; and a communication part 709 including a network interface card such as a LAN card, a modem, etc. The communication part 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as necessary. A removable medium 711 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 710 as necessary, so that a computer program read out therefrom is installed in the storage part 708 as necessary.

[0158] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program according to embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for executing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication part 709, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit (CPU) 701, the above-described functions defined in the system of the present disclosure are executed.

[0159] It should be noted that the computer-readable medium shown in the present application can 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, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, 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 application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or component. In the present application, the computer-readable signal medium can include a data signal carried in a baseband or as a part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, which can send, propagate or transmit a program for use by or in conjunction with an instruction execution system, device or component. The program code contained on the computer-readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0160] The flowcharts and block diagrams in the drawings illustrate the possible implementation architectures, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment or a part of code containing one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different order than that shown in the drawings. For example, two blocks that are shown in succession can actually be executed substantially in parallel, and sometimes in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams or flowcharts, and the combination of blocks in the block diagrams or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0161] The modules described in the embodiments of the present application can be implemented in the form of software or hardware. The described modules can also be arranged in a processor, for example, a processor can be described as including a query module, a hard matching module, a similarity matching module, and an output module. In some cases, the names of these modules do not constitute a limitation on the modules themselves, for example, the query module can also be described as a "standard product attribute query module".

[0162] As another aspect, the present application also provides a computer readable medium, which can be included in the device described in the above embodiments, or can exist independently without being assembled into the device. The computer readable medium carries one or more programs, which, when executed by the device, cause the device to include:

[0163] Collecting information related to a first product and a second product respectively, and querying whether the first product and the second product both belong to a standard product; wherein the first product and the second product belong to different platforms;

[0164] If both belong to the standard product, performing product category normalization processing on the information of the first product and the second product by using a hard matching algorithm; or

[0165] If both do not belong to the standard product, performing product category normalization processing on the information of the first product and the second product by using a similarity matching algorithm;

[0166] Outputting a processing result of whether the first product and the second product belong to the same product category or different product categories.

[0167] According to the technical solution of the embodiments of the present application, the information of products collected by multiple platforms is processed by using a hard matching algorithm or a similarity matching algorithm according to whether the products both belong to a standard product, so that products belonging to the same SKU are classified into the same category, and the purpose of cross-platform product category normalization is achieved, thereby realizing the standardization of product category management of different platforms.

[0168] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. An item category normalization method applied across platforms, characterized by, The method comprises the following steps: Collecting information related to the first and second items respectively, and querying whether the first and second items belong to a standard item; wherein the first and second items belong to different platforms; If both belong to the standard item, using a hard matching algorithm to perform category normalization processing on the information of the first and second items; or If only one of the first and second items belongs to the standard item, determining that the first and second items belong to different categories; If neither belongs to the standard item, using a similarity matching algorithm to perform category normalization processing on the information of the first and second items, comprising: for any one of the first and second items, using a title cutting comparison method to cut the title of the any one item; if there is no brand name in the cut word, adding the brand name of the any one item to the title of the any one item; calculating the similarity of the titles of the first and second items, if the similarity is less than a first preset similarity threshold, determining that the first and second items belong to different categories; otherwise, calculating the similarity of the model numbers of the first and second items, if the similarity is greater than or equal to a second preset similarity threshold, determining that the first and second items belong to the same category; if the similarity is less than the second preset similarity threshold, calculating the difference between the values of the first and second items; if the difference is within a preset difference range, determining that the first and second items belong to the same category, otherwise, belonging to different categories; Outputting the processing result of whether the first and second items belong to the same category or different categories.

2. The method of claim 1, wherein, The item information includes a brand and a specification parameter, and the specification parameter includes a model number; The category normalization processing on the information of the first and second items using the hard matching algorithm comprises: Determining whether the model numbers and the brands of the first and second items are the same, if both are the same, determining that the first and second items belong to the same category, otherwise, belonging to different categories.

3. The method of claim 2, wherein, The first and second items are electronic products, and the specification parameter further includes a memory model ROM value; The determination of whether the model numbers and the brands of the first and second items are the same comprises: Determining whether the model numbers, ROM values and brands of the first and second items are the same.

4. The method of claim 2, wherein, The item information includes a title; Before the determination of whether the model numbers and the brands of the first and second items are the same, the method further comprises: For any one of the first and second items, if there is no model number in the specification parameter, obtaining the model number from the title of the any one item by cutting; and If the model numbers of the first and second items are in different formats, performing normalization processing on the model numbers of the first and second items according to a preset model number format.

5. The method of claim 1, wherein, The calculation of the similarity of the titles of the first and second items comprises: extracting a first feature from a title of the first item, generating a first text feature vector based on the first feature; and extracting a second feature from a title of the second item, generating a second text feature vector based on the second feature; calculating a similarity of the first text feature vector and the second text feature vector.

6. An item category normalizing device applied to cross-platforms, characterized by, Comprising: a query module, configured to collect information related to a first item and a second item respectively, and query whether the first item and the second item both belong to a standard item; wherein the first item and the second item belong to different platforms; a hard matching module, configured to, if both belong to the standard item, perform category normalization processing on information of the first item and the second item by using a hard matching algorithm; or a similarity matching module, configured to, if only one of the first item and the second item belongs to the standard item, determine that the first item and the second item belong to different categories; if neither belongs to the standard item, perform category normalization processing on information of the first item and the second item by using a similarity matching algorithm, including: for any one of the first item and the second item, performing word segmentation processing on a title of the any one item by using a title word segmentation comparison method; if there is no brand name in the segmented words obtained by word segmentation, adding a brand name to which the any one item belongs to the title of the any one item; calculating a similarity of the title of the first item and the title of the second item, if the similarity is less than a first preset similarity threshold, determining that the first item and the second item belong to different categories; otherwise, calculating a similarity of a model of the first item and a model of the second item, if the similarity is greater than or equal to a second preset similarity threshold, determining that the first item and the second item belong to the same category; if the similarity is less than the second preset similarity threshold, calculating a difference value of a value of the first item and a value of the second item; if the difference value is within a preset difference value range, determining that the first item and the second item belong to the same category, otherwise, belong to different categories; an output module, configured to output a processing result of the first item and the second item belonging to the same category or different categories.

7. An electronic device, comprising: Comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1-5.

8. A computer readable medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the method of any one of claims 1-5.

Images