A cuisine calibration method, device, electronic device and non-volatile storage medium
The vector distance between merchants and cuisines is calculated by vector determination model, which solves the problem of low calibration accuracy of cuisines in the prior art, and realizes automatic and accurate calibration of merchants' cuisines.
Patent Information
- Application Number
- CN201910271616.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-04-04
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2039-04-04
AI Technical Summary
In the prior art, merchant cuisine calibration is prone to negligence and misjudgment, which affects the accuracy of cuisine calibration, mainly because the cuisine structure is complex and relies on manual judgment.
The vector determination model is used to obtain the vector information of the merchant and the cuisine to be calibrated. By calculating the distance between the two, the cuisine is calibrated according to the distance set to ensure the accuracy of the calibration. The specific steps include obtaining text information of the merchant and the cuisine, performing word segmentation and segmentation processing, calculating cosine similarity to obtain a distance set, and determining the cuisine type based on the preset threshold.
It realizes automatic calibration of merchant cuisine, improves the accuracy and efficiency of cuisine calibration, and reduces the possibility of manual misjudgment.
Smart Images

Figure CN111858838B_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present invention relate to the field of communication technology, and in particular to a cuisine calibration method, device, electronic device and non-volatile storage medium. Background Art
[0002] In the field of restaurant takeout, the cuisines served by merchants usually include: Shandong cuisine, Sichuan cuisine, Cantonese cuisine, Jiangsu cuisine, Zhejiang cuisine, Fujian cuisine, Hunan cuisine and Anhui cuisine, etc. If merchants indicate the corresponding cuisine type, it can often significantly affect users' choices when ordering online.
[0003] The inventors have discovered that there are at least the following problems in the related art: In the prior art, manual judgment is usually used to calibrate a merchant's cuisine. However, due to the complexity of the cuisine structure, if calibration is performed only through manual judgment, negligence and misjudgment are often prone to occur, thereby affecting the accuracy of the merchant's cuisine calibration. Summary of the invention
[0004] The purpose of the embodiments of the present invention is to provide a cuisine calibration method, device, electronic device and non-volatile storage medium, so as to accurately realize the automatic calibration of merchant cuisine.
[0005] To solve the above technical problems, an embodiment of the present invention provides a cuisine calibration method, which is applied to an electronic device and includes the following steps: inputting the text information of a merchant to be calibrated into a vector determination model to obtain the vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the lexical features of the text information of the merchant to be calibrated; inputting the text information of a cuisine into a vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the lexical features of the text information of the cuisine, and the number of cuisines is at least two; calculating the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine and obtaining a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine; and calibrating the cuisine of the merchant to be calibrated according to the distance set.
[0006] An embodiment of the present invention also provides a cuisine calibration device, including: a first determination module, used to input the text information of the merchant to be calibrated into a vector determination model to obtain the vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the lexical features of the text information of the merchant to be calibrated; a second determination module, used to input the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the lexical features of the text information of the cuisine, and the number of cuisines is at least two; a distance calculation module, used to calculate the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine and obtain a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine; a cuisine calibration module, used to perform cuisine calibration on the merchant to be calibrated according to the distance set.
[0007] An embodiment of the present invention also provides an electronic device, including a memory and a processor, the memory storing a computer program, and the processor executing the program: inputting the text information of the merchant to be calibrated into a vector determination model to obtain the vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the lexical features of the text information of the merchant to be calibrated; inputting the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the lexical features of the text information of the cuisine, and the number of cuisines is at least two; calculating the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine and obtaining a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine; and calibrating the cuisine of the merchant to be calibrated according to the distance set.
[0008] An embodiment of the present invention further provides a non-volatile storage medium for storing a computer-readable program, wherein the computer-readable program is used for a computer to execute the above-mentioned method for calibrating a dish.
[0009] Compared with the prior art, the main difference and effect of the embodiments of the present invention are: it is possible to obtain the vector information of the merchant to be calibrated and the vector information of the cuisine through a vector determination model, and determine the cuisine that needs to be calibrated by the merchant based on the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine, thereby accurately realizing automatic calibration of the merchant's cuisine.
[0010] In addition, the cuisine calibration of the merchant to be calibrated according to the distance set includes: determining at least one cuisine corresponding to a distance less than a preset threshold in the distance set; and calibrating the cuisine of the merchant to be calibrated according to the at least one cuisine. In this implementation, the at least one cuisine corresponding to a distance less than a preset threshold in the distance set is used to calibrate the merchant to be calibrated, so that the cuisine calibrated by the merchant is more accurate.
[0011] In addition, before inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, it also includes: obtaining the description information of the sample merchant, wherein the number of the sample merchants is at least two; obtaining the sample text information according to the description information of the sample merchant, wherein the sample text information includes at least two sample words; calculating the vector of each sample word in the sample text information according to the glove algorithm; training according to each sample word in the sample text information and the vector of each sample word to obtain the vector determination model, wherein the vector determination model includes a corresponding relationship between the storage location identifier for representing the word and the word vector. In this implementation, the sample text information is obtained by processing the description information of the sample merchant, the vector of each sample word in the sample text information is obtained by the glove algorithm, and the vector determination model is obtained by the obtained sample words and the vector of the sample words, thereby ensuring the accuracy of the vector determination model.
[0012] In addition, sample text information is obtained based on the description information of the sample merchants, including: performing word segmentation processing on the description information of the sample merchants to obtain sample segmentation results; performing single-character segmentation processing on the description information of the sample merchants to obtain sample segmentation results; and sequentially connecting the sample word segmentation results and the sample segmentation results to obtain sample text information.
[0013] In addition, before inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, it also includes: obtaining the description information of the merchant to be calibrated; performing word segmentation processing on the description information of the merchant to be calibrated to obtain a word segmentation result; performing single-character segmentation on the description information of the merchant to be calibrated to obtain a segmentation result; and sequentially connecting the word segmentation result and the segmentation result to obtain the text information of the merchant to be calibrated. In this implementation, by performing word segmentation processing and single-character segmentation processing on the description information of the merchant to be calibrated, and obtaining the text information of the merchant to be calibrated according to the word segmentation results and segmentation results respectively obtained, it is ensured that the obtained text information of the merchant to be calibrated consists of multiple words related to the merchant.
[0014] In addition, the text information of the merchant to be calibrated is input into the vector determination model to obtain the vector information of the merchant to be calibrated, including: determining the merchant vocabulary contained in the text information of the merchant to be calibrated; obtaining the vector of each merchant vocabulary in the text information of the merchant to be calibrated according to the merchant vocabulary and the vector determination model; adding and averaging the vectors of each merchant vocabulary in the text information of the merchant to be calibrated to obtain the vector information of the merchant to be calibrated. In this implementation, by obtaining the vector of each merchant vocabulary in the text information of the merchant to be calibrated, and obtaining the vector information of the merchant to be calibrated by adding and averaging the vectors of each merchant vocabulary, the obtained vector information of the merchant to be calibrated is made more accurate.
[0015] In addition, the text information of the cuisine is input into the vector determination model to obtain the vector information of the cuisine, including: determining the cuisine vocabulary corresponding to the text information of the cuisine; obtaining the vector of the cuisine vocabulary according to the cuisine vocabulary and the vector determination model; and using the vector of the cuisine vocabulary as the vector information of the cuisine.
[0016] In addition, the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine is calculated and a distance set is obtained, including: calculating the cosine similarity between the vector information of the merchant to be calibrated and the vector information of each cuisine; obtaining the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine according to the cosine similarity; and obtaining a distance set according to the distance.
[0017] In addition, the description information of the merchant to be marked includes: the name of the merchant to be marked, the product name of the merchant to be marked, and the user evaluation of the merchant to be marked. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.
[0019] Figure 1 is a flow chart of a method for labeling a cuisine in a first embodiment of the present application;
[0020] Figure 2 is a flow chart of a method for labeling a cuisine in a second embodiment of the present application;
[0021] Figure 3 This is a schematic diagram of a cuisine calibration device in the third embodiment of the present application;
[0022] Figure 4 This is a schematic diagram of a cuisine calibration device in the fourth embodiment of the present application;
[0023] Figure 5 It is a schematic diagram of the structure of an electronic device in the fifth embodiment of the present application. DETAILED DESCRIPTION
[0024] To make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. However, it will be appreciated by those skilled in the art that in the embodiments of the present invention, many technical details are proposed in order to enable the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present application can be implemented. The division of the following embodiments is for the convenience of description, and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined and referenced with each other without contradiction.
[0025] The first embodiment of the present invention relates to a method for calibrating a cuisine, which is applied to an electronic device. The flowchart of the method can be as follows: Figure 1 As shown, specifically including:
[0026] Step 101: Input text information of the merchant to be calibrated into a vector determination model to obtain vector information of the merchant to be calibrated.
[0027] Specifically, the vector information of the merchant to be calibrated represents the lexical features of the text information of the merchant to be calibrated, and before the text information of the merchant to be calibrated is input into the vector determination model, the text information of the merchant to be calibrated needs to be obtained first. And in this embodiment, the method used to obtain the text information of the merchant to be calibrated is specifically: obtain the description information of the merchant to be calibrated, perform word segmentation processing on the description information of the merchant to be calibrated to obtain the word segmentation result, perform single-character segmentation on the description information of the merchant to be calibrated to obtain the segmentation result, and sequentially connect the word segmentation result and the segmentation result to obtain the text information of the merchant to be calibrated. And in this embodiment, the description information of the merchant to be calibrated includes the name of the merchant to be calibrated, the product name of the merchant to be calibrated, and the user evaluation of the merchant to be calibrated.
[0028] In a specific implementation, the description information of the merchant to be calibrated includes "Wangxiang Garden Xiangxi Overlord Frog is delicious", where "Wangxiang Garden" represents the name of the merchant to be calibrated, "Xiangxi Overlord Frog" represents the product name of the merchant to be calibrated, and "very delicious" represents the user evaluation of the merchant to be calibrated. The description information of the merchant to be calibrated is segmented to obtain the segmentation result of "Wangxiang Garden Xiangxi Overlord Frog is delicious", and the description information of the merchant to be calibrated is segmented by single characters to obtain the segmentation result of "Wangxiang Garden Xiangxi Overlord Frog is delicious", and the segmentation result and the segmentation result are sequentially connected to obtain "Wangxiang Garden Xiangxi Overlord Frog is delicious Wangxiang Garden Xiangxi Overlord Frog is delicious", and the repeated words in the obtained result are deleted to obtain the text information of the merchant to be calibrated as "Wangxiang Garden Xiangxi Overlord Frog is delicious Wangxiang Garden West Overlord Frog is delicious".
[0029] It should be noted that in this embodiment, only an example is given where the description information of the merchant to be calibrated contains one commodity and user evaluation words. In actual applications, the description information of the merchant to be calibrated may also include multiple commodities and multiple user evaluation words. For example, the description information of the merchant to be calibrated may also include "Looking forward to Western Hunan's Western Hunan Overbearing Frog and Hunan-flavored stir-fries are very delicious, spicy, tender and soft". Among them, "Western Hunan Overbearing Frog" and "Hunan-flavored stir-fries" respectively represent the commodity names of the merchant to be calibrated, and "very delicious" and "spicy, tender and soft" respectively represent the user evaluation words of the merchant to be calibrated. And when the description information of the merchant to be calibrated includes multiple commodities and multiple user evaluation words, the method of obtaining the text information of the merchant to be calibrated is roughly the same as the method of obtaining the text information of the merchant to be calibrated that only includes one commodity and user evaluation words, except that the content of the obtained text information of the merchant to be calibrated is richer and the cuisine calibration is more accurate. Therefore, the number of commodities and the number of user evaluation words in the description information of the merchant to be calibrated can be determined according to the actual needs of the user, and this will not be elaborated in this embodiment.
[0030] Specifically, in this embodiment, the vector information of the merchant to be calibrated can be obtained by using the obtained text information of the merchant to be calibrated. The specific method adopted can be to determine the merchant vocabulary contained in the text information of the merchant to be calibrated, and according to the merchant vocabulary and the vector determination model, obtain the vectors of each merchant vocabulary in the text information of the merchant to be calibrated, where the vector determination model contains the corresponding relationship between the vocabulary and the storage location identifier of the vocabulary vector. Perform an addition and averaging operation on the vectors of each merchant vocabulary in the text information of the merchant to be calibrated to obtain the vector information of the merchant to be calibrated. For example, multiple sample vocabularies are stored in the vector determination model, and each sample vocabulary is respectively associated with the storage space of its vector through an identifier. For example, assuming the sample vocabulary is "Looking forward to Western Hunan", the identifier of the vector storage space where the vector of the vocabulary "Looking forward to Western Hunan" is stored in the vector determination model is 1, and the vector stored in the vector storage space with the identifier 1 is [0.65 0.14]. Of course, only the example corresponding to the identifier 1 of "Looking forward to Western Hunan" is given in this embodiment. The principle of the storage location identifier of the vocabulary vector corresponding to other sample vocabularies is roughly the same, and this will not be elaborated in this embodiment.
[0031] In a specific implementation, the text information of the merchant to be calibrated is determined to be "Wangxiangyuan Xiangxi Overlord Frog is very delicious Wangxiangyuan West Overlord Frog is very delicious", and the merchant vocabulary contained in the text information of the merchant to be calibrated includes "Wangxiangyuan", "Xiangxi", "Overlord", "Frog", "Very delicious", "Wang", "Xiang", "Garden", "West", "Overlord", "King", "Frog", "Very", "Beautiful", and "Taste". The merchant vocabulary "Wangxiangyuan" is input into the vector determination model, and the vector determination model can determine that the identifier corresponding to the vocabulary "Wangxiangyuan" is 1, and obtain the vector [0.65 0.14] from the vector storage space identified as 1. In the same way, the remaining words in the text information of the merchant to be calibrated are input into the vector determination model respectively, and the vectors of the remaining words can be obtained as follows: Xiangxi: [0.610.62] Overlord: [0.31 0.93] Frog: [0.10 0.21] Very delicious: [0.43 0.01] Wang: [0.05 0.54] Xiang: [0.720.13] Garden: [0.54 0.68] West: [0.98 0.18] Overlord: [0.86 0.84] King: [0.29 0.56] Frog: [0.70 0.34] Very: [0.24 0.19] Beautiful: [0.23 0.49] Taste: [0.54 0.05], the vectors of each merchant word in the text information of the merchant to be calibrated obtained above are added and averaged, and the vector information of the merchant to be calibrated is obtained as X=[0.48 0.39].
[0032] Step 102, input the text information of the cuisine, and the vector determination model obtains the vector information of the cuisine.
[0033] Specifically, in this embodiment, the vector information of the cuisine represents the lexical features of the text information of the cuisine, and the number of cuisines is at least two. The types of cuisines are known, for example, including: Shandong cuisine, Sichuan cuisine, Cantonese cuisine, Jiangsu cuisine, Zhejiang cuisine, Fujian cuisine, Hunan cuisine or Anhui cuisine. And the cuisine vocabulary corresponding to the text information of the cuisine is determined to be: Shandong, Sichuan, Guangdong, Jiangsu, Zhejiang, Fujian, Hunan or Anhui. Each cuisine vocabulary is input into the vector determination model respectively, and the vector determination model can determine the identifier of the storage location corresponding to each cuisine vocabulary, and determine the storage location from the obtained identifier, and obtain the vector of each cuisine from the storage location. For example, the vectors of the cuisine vocabulary obtained are Shandong: [0.33 0.12] Sichuan: [0.22 0.08] Guangdong: [0.34 0.67] Jiangsu: [0.98 0.34] Zhejiang: [0.11 0.25] Fujian: [0.67 0.45] Hunan: [0.67 0.82] Anhui: [0.560.67], and the vectors of the cuisine vocabulary obtained are used as the vector information of the cuisine.
[0034] It should be noted that this embodiment is only described by way of example, and in actual application, the type of marked cuisine can be determined as needed, and this embodiment is not limited to the types of cuisine listed in the above embodiments of the present application.
[0035] Step 103, calculate the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine and obtain a distance set.
[0036] Specifically, the cosine similarity between the vector information of the merchant to be calibrated and the vector information of each cuisine is calculated, and the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine is obtained according to the cosine similarity. A distance set is obtained according to the distance, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine.
[0037] In a specific implementation, the vector information of the merchant to be calibrated is determined to be X=[0.48 0.39], while the vector information of Shandong cuisine is A=[0.33 0.12], the vector information of Sichuan cuisine is B=[0.22 0.08], the vector information of Cantonese cuisine is C=[0.340.67], the vector information of Jiangsu cuisine is D=[0.98 0.34], the vector information of Zhejiang cuisine is E=[0.11 0.25], the vector information of Fujian cuisine is F=[0.67 0.45], the vector information of Hunan cuisine is G=[0.67 0.82], and the vector information of Anhui cuisine is H=[0.560.67]. Taking Shandong cuisine as an example, the distance between the vector information X of the merchant to be calibrated and the vector information A of Shandong cuisine is calculated: Among them, the cosine similarity between the merchant's vector information X and the vector information A of Shandong cuisine is 78 degrees, and the cosine distance is 0.208. Using the same calculation principle, we can obtain: the cosine similarity between the merchant's vector information X and the vector information B of Sichuan cuisine is 65 degrees, and the cosine distance is 0.422; the cosine similarity between the merchant's vector information X and the vector information C of Cantonese cuisine is 58 degrees, and the cosine distance is 0.530; the cosine similarity between the merchant's vector information X and the vector information D of Jiangsu cuisine is 50 degrees, and the cosine distance is 0. The chord distance is 0.642; the cosine similarity between the merchant's vector information X and the Zhejiang cuisine's vector information E is 48, and the cosine distance is 0.669; the cosine similarity between the merchant's vector information X and the Fujian cuisine's vector information F is 35 degrees, and the cosine distance is 0.819; the cosine similarity between the merchant's vector information X and the Hunan cuisine's vector information G is 15 degrees, and the cosine distance is 0.965; the cosine similarity between the merchant's vector information X and the Anhui cuisine's vector information H is 37 degrees, and the cosine distance is 0.789. Since the smaller the cosine similarity, the larger the cosine distance, the more similar the two vectors are, and the smaller the actual distance is. Therefore, the cosine distance can be determined based on the cosine similarity, and the inverse of the cosine distance is used as the actual distance between the two vectors, so that the distance set obtained is {d XA =1000 / 208d XB =1000 / 422d XC =1000 / 530d XD =1000 / 642d XE =1000 / 669d XF =1000 / 819d XG =1000 / 965d XH =1000 / 789}, where d XA It represents the actual distance between the vector information X of the merchant and the vector information A of Shandong cuisine. The meanings of other symbols are roughly the same and will not be described in detail in the implementation manner of this application.
[0038] Step 104, calibrate the cuisine of the merchant to be calibrated according to the distance set.
[0039] Specifically, in an embodiment of the present application, at least one cuisine corresponding to a distance in a distance set that is less than a preset threshold is determined, and cuisine calibration is performed on the merchant to be calibrated based on the at least one cuisine.
[0040] For example, the determined distance set is {d XA =1000 / 208d XB =1000 / 422d XC =1000 / 530d XD =1000 / 642d XE =1000 / 669dXF =1000 / 819d XG =1000 / 965d XH =1000 / 789}, and the preset threshold distance is 1000 / 900, and the preset threshold in this embodiment can be set by the user according to actual needs, and of course it can also be a value set according to historical calibration records through statistics. The specific acquisition method of the preset threshold is not limited in this embodiment. Therefore, it can be determined that the distance less than the preset threshold in the distance set is 1000 / 965, and the cuisine corresponding to this distance is G, that is, Hunan cuisine. Therefore, it can be determined that the cuisine type corresponding to the merchant "Wang Xiangyuan" to be calibrated is Hunan cuisine, so the merchant "Wang Xiangyuan" to be calibrated is marked as Hunan cuisine. Of course, in this embodiment, only one cuisine exceeds the preset threshold for illustration, and there may be multiple cuisines that exceed the preset threshold. In actual applications, multiple cuisines that exceed the preset threshold can be calibrated, and it is explained that the goods sold in the merchant to be calibrated are not a single cuisine, but contain multiple cuisines.
[0041] Compared with the prior art, the cuisine calibration method provided in this embodiment can obtain the vector information of the merchant to be calibrated and the vector information of the cuisine through a vector determination model, and determine the cuisine that the merchant needs to calibrate based on the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine, thereby accurately realizing automatic calibration of the merchant's cuisine.
[0042] The second embodiment of the present invention relates to a method for calibrating a cuisine. This embodiment is further improved on the basis of the first embodiment. The specific improvement is that before inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, a step of obtaining a vector determination model is added. The process of the information transmission method in this embodiment is as follows: Figure 2 Specifically, in this embodiment, steps 201 to 205 are included, wherein steps 204 to 205 are substantially the same as steps 104 to 105 in the first embodiment, and will not be described in detail here. The following mainly introduces the differences. For technical details not described in detail in this embodiment, please refer to the cuisine calibration method provided in the first embodiment, and will not be described in detail here.
[0043] Step 201, obtaining a vector determination model.
[0044] Specifically, in this embodiment, the method of obtaining the vector determination model can be specifically adopted: obtaining the description information of the sample merchants, and the number of the sample merchants is at least two, obtaining the sample text information according to the description information of the sample merchants, and the sample text information includes at least two sample words, and calculating the vector of each sample word in the sample text information according to the glove algorithm, wherein the glove algorithm uses the frequency of co-occurrence of the test word and the words in the corpus to calculate the vector information of the test word, and the words in the corpus are very large, so when the test word is a sample word, the vector of each sample word can be calculated according to the frequency of the sample word and the words that co-occur in the corpus. Since the prior art of the glove algorithm is relatively mature and the glove algorithm is not the focus of this application, it will not be repeated in the implementation of this application.
[0045] Among them, training is performed according to each sample word and the vector of each sample word in the sample text information to obtain a vector determination model, and the vector determination model includes a corresponding relationship between the storage location identifier for representing the word and the word vector. For example, if the sample word is "very delicious", then a matching relationship between "very delicious" and the vector storage space identified as 1, that is, the storage location, is established, and the vector stored in the vector storage space identified as 1 is [0.65 0.14]. Therefore, as long as any word is input, the vector determination model can determine the identifier of the storage location corresponding to the word, and obtain the vector corresponding to the word through the storage location corresponding to the identifier. By establishing a corresponding relationship between the storage location identifier of the word and the word vector, the storage space of the vector determination model can be further saved, thereby improving the efficiency of obtaining the word vector. Of course, this embodiment is explained by taking a sample word as an example, and in actual applications, the vector determination model almost covers all words in the field of online food ordering.
[0046] Among them, when obtaining sample text information based on the description information of the sample merchants, the sample merchant description information is specifically processed by word segmentation to obtain a sample segmentation result, the sample merchant description information is processed by single-character segmentation to obtain a sample segmentation result, and the sample word segmentation result and the sample segmentation result are sequentially connected to obtain the sample text information.
[0047] It should be noted that the method of obtaining sample text information in the implementation mode of the present application is substantially the same as the method of obtaining text information of the merchant to be calibrated in the first implementation mode, and therefore will not be described in detail in the implementation mode of the present application.
[0048] After step 201, steps 202 to 205 are executed.
[0049] Compared with the prior art, the cuisine calibration method provided in this embodiment can obtain the vector information of the merchant to be calibrated and the vector information of the cuisine through the vector determination model, and determine the cuisine that the merchant needs to calibrate according to the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine, so as to accurately realize the automatic calibration of the merchant's cuisine. And the vector determination model is obtained through the sample words obtained in the description information of the sample merchant and the vectors of the sample words, so as to ensure the accuracy of the vector determination model.
[0050] The step division of the above methods is only for the purpose of clear description. When implemented, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this patent; adding insignificant modifications to the algorithm or process or introducing insignificant designs without changing the core design of the algorithm and process are all within the scope of protection of this patent.
[0051] A third embodiment of the present invention relates to a cuisine calibration device, such as Figure 3 As shown, the device includes: a first determination module 31, a second determination module 32, a distance calculation module 33 and a cuisine calibration module 34.
[0052] The first determination module 31 is used to input the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the lexical features of the text information of the merchant to be calibrated.
[0053] The second determination module 32 is used to input the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the vocabulary features of the text information of the cuisine, and the number of the cuisines is at least two.
[0054] The distance calculation module 33 is used to calculate the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine and obtain a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine.
[0055] The cuisine calibration module 34 is used to calibrate the cuisine of the merchants to be calibrated according to the distance set.
[0056] It is not difficult to find that this embodiment is a device embodiment corresponding to the first embodiment, and this embodiment can be implemented in conjunction with the first embodiment. The relevant technical details mentioned in the first embodiment are still valid in this embodiment, and in order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied in the first embodiment.
[0057] A fourth embodiment of the present invention relates to a cuisine calibration device, such as Figure 4 The diagram is a schematic diagram of the cuisine calibration device in this embodiment. This embodiment is further improved on the basis of the third embodiment, and the specific improvement is the addition of a model acquisition module 30.
[0058] The model acquisition module 30 is used to obtain a vector determination model.
[0059] The first determination module 31 is used to input the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the lexical features of the text information of the merchant to be calibrated.
[0060] The second determination module 32 is used to input the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the vocabulary features of the text information of the cuisine, and the number of the cuisines is at least two.
[0061] The distance calculation module 33 is used to calculate the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine and obtain a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine.
[0062] The cuisine calibration module 34 is used to calibrate the cuisine of the merchants to be calibrated according to the distance set.
[0063] It is not difficult to find that this embodiment is a device embodiment corresponding to the second embodiment, and this embodiment can be implemented in conjunction with the second embodiment. The relevant technical details mentioned in the second embodiment are still valid in this embodiment, and in order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied in the second embodiment.
[0064] A fifth embodiment of the present invention relates to an electronic device, such as Figure 5 As shown, the electronic device includes: at least one processor 501; and a memory 502 connected to the at least one processor 501 in communication; wherein the memory 502 stores instructions that can be executed by the at least one processor 501, and the instructions are executed by the at least one processor 501 to implement:
[0065] Input the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the lexical features of the text information of the merchant to be calibrated; input the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the lexical features of the text information of the cuisine, and the number of cuisines is at least two; calculate the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine and obtain a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine; calibrate the cuisine of the merchant to be calibrated according to the distance set.
[0066] Specifically, the electronic device includes: one or more processors 501 and a memory 502, Figure 5 A processor 501 is taken as an example. The processor 501 and the memory 502 may be connected via a bus or other means. Figure 5 In the example, the bus connection is used. The memory 502 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules. The processor 501 executes various functional applications and data processing of the device by running the non-volatile software programs, instructions and modules stored in the memory 502, that is, realizing the above-mentioned cuisine calibration method.
[0067] The memory 502 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store a list of options, etc. In addition, the memory 502 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 502 may optionally include a memory 502 remotely arranged relative to the processor 501, and these remote memories 502 may be connected to an external device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0068] One or more modules are stored in the memory 502, and when executed by one or more processors 501, the cuisine calibration method in any of the above method implementations is executed.
[0069] The above-mentioned product can execute the method provided in the implementation mode of this application, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this implementation mode, please refer to the cuisine calibration method provided in the implementation mode of this application.
[0070] The sixth embodiment of the present application relates to a non-volatile storage medium, in which a computer program is stored. The computer-readable program is used for a computer to execute the cuisine calibration method involved in the first or second embodiment of the present invention.
[0071] Those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including several instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0072] Those skilled in the art will appreciate that the above-mentioned embodiments are specific examples for implementing the present invention, and in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present invention.
[0073] The present application embodiment discloses A1. a method for calibrating a cuisine, comprising:
[0074] Inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the lexical features of the text information of the merchant to be calibrated;
[0075] Inputting the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the vocabulary features of the text information of the cuisine, and the number of the cuisines is at least two;
[0076] Calculating the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines and obtaining a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines;
[0077] The cuisine of the merchant to be calibrated is calibrated according to the distance set.
[0078] A2. According to the cuisine calibration method described in A1, the cuisine calibration of the merchant to be calibrated according to the distance set includes:
[0079] Determine at least one cuisine corresponding to a distance in the distance set that is less than a preset threshold;
[0080] The merchant to be calibrated is calibrated according to the at least one cuisine.
[0081] A3. The method for calibrating a cuisine according to A1, before inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, further comprising:
[0082] Obtaining description information of sample merchants, wherein the number of the sample merchants is at least two;
[0083] Obtaining sample text information according to the description information of the sample merchant, wherein the sample text information includes at least two sample words;
[0084] Calculate the vector of each sample word in the sample text information according to the glove algorithm;
[0085] The vector determination model is obtained by training the sample words and the vectors of the sample words in the sample text information, wherein the vector determination model includes a corresponding relationship between storage location identifiers for representing words and word vectors.
[0086] A4. According to the cuisine calibration method described in A3, the sample text information is obtained according to the description information of the sample merchant, including:
[0087] Performing word segmentation processing on the description information of the sample merchant to obtain a sample word segmentation result;
[0088] Performing single-character segmentation processing on the description information of the sample merchant to obtain a sample segmentation result;
[0089] The sample word segmentation result and the sample segmentation result are sequentially connected to obtain the sample text information.
[0090] A5. The method for calibrating a cuisine according to A4, wherein before inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, the method further comprises:
[0091] Obtaining description information of the merchant to be calibrated;
[0092] Performing word segmentation processing on the description information of the merchant to be calibrated to obtain a word segmentation result;
[0093] Performing single-character segmentation on the description information of the merchant to be calibrated to obtain a segmentation result;
[0094] The word segmentation result and the segmentation result are sequentially connected to obtain the text information of the merchant to be marked.
[0095] A6. The method for calibrating a cuisine according to A5, wherein the step of inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated comprises:
[0096] Determining the merchant vocabulary contained in the text information of the merchant to be marked;
[0097] According to the merchant vocabulary and the vector determination model, obtaining the vector of each merchant vocabulary in the text information of the merchant to be calibrated;
[0098] The vectors of each merchant word in the text information of the merchant to be calibrated are added and averaged to obtain the vector information of the merchant to be calibrated.
[0099] A7. The method for calibrating a cuisine according to A4, wherein the step of inputting the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine comprises:
[0100] Determining the cuisine vocabulary corresponding to the text information of the cuisine;
[0101] Determine a model according to the cuisine vocabulary and the vector, and obtain the vector of the cuisine vocabulary;
[0102] The vector of the cuisine vocabulary is used as the vector information of the cuisine.
[0103] A8. According to the cuisine calibration method described in A1, the step of calculating the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine and obtaining a distance set includes:
[0104] Calculate the cosine similarity between the vector information of the merchant to be calibrated and the vector information of each cuisine;
[0105] Obtaining the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines according to the cosine similarity;
[0106] The distance set is obtained according to the distance.
[0107] A9. According to the cuisine labeling method described in A5 to A8, the description information of the merchant to be labeled includes: the name of the merchant to be labeled, the product name of the merchant to be labeled, and the user evaluation of the merchant to be labeled.
[0108] The present application embodiment discloses B1. a cuisine calibration device, comprising:
[0109] A first determination module is used to input the text information of the merchant to be calibrated into a vector determination model to obtain vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the lexical features of the text information of the merchant to be calibrated;
[0110] A second determination module is used to input the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the vocabulary features of the text information of the cuisine, and the number of the cuisines is at least two;
[0111] A distance calculation module, used to calculate the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines and obtain a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines;
[0112] The cuisine calibration module is used to calibrate the cuisine of the merchant to be calibrated according to the distance set.
[0113] The embodiment of the present application discloses C1. An electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes:
[0114] Inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the lexical features of the text information of the merchant to be calibrated;
[0115] Inputting the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the vocabulary features of the text information of the cuisine, and the number of the cuisines is at least two;
[0116] Calculating the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines and obtaining a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines;
[0117] The cuisine of the merchant to be calibrated is calibrated according to the distance set.
[0118] C2. The electronic device according to C1, wherein the step of performing cuisine calibration on the merchant to be calibrated according to the distance set comprises:
[0119] Determine at least one cuisine corresponding to a distance in the distance set that is less than a preset threshold;
[0120] The merchant to be calibrated is calibrated according to the at least one cuisine.
[0121] C3. The electronic device according to C1, before inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, further comprising:
[0122] Obtaining description information of sample merchants, wherein the number of the sample merchants is at least two;
[0123] Obtaining sample text information according to the description information of the sample merchant, wherein the sample text information includes at least two sample words;
[0124] Calculate the vector of each sample word in the sample text information according to the glove algorithm;
[0125] The vector determination model is obtained by training the sample words and the vectors of the sample words in the sample text information, wherein the vector determination model includes a corresponding relationship between storage location identifiers for representing words and word vectors.
[0126] C4. The electronic device according to C3, wherein obtaining the sample text information according to the description information of the sample merchant comprises:
[0127] Performing word segmentation processing on the description information of the sample merchant to obtain a sample word segmentation result;
[0128] Performing single-character segmentation processing on the description information of the sample merchant to obtain a sample segmentation result;
[0129] The sample word segmentation result and the sample segmentation result are sequentially connected to obtain the sample text information.
[0130] C5. The electronic device according to C4, before inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, further comprising:
[0131] Obtaining description information of the merchant to be calibrated;
[0132] Performing word segmentation processing on the description information of the merchant to be calibrated to obtain a word segmentation result;
[0133] Performing single-character segmentation on the description information of the merchant to be calibrated to obtain a segmentation result;
[0134] The word segmentation result and the segmentation result are sequentially connected to obtain the text information of the merchant to be marked.
[0135] C6. The electronic device according to C5, wherein the step of inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated comprises:
[0136] Determining the merchant vocabulary contained in the text information of the merchant to be marked;
[0137] According to the merchant vocabulary and the vector determination model, obtaining a vector of each merchant vocabulary in the first text;
[0138] The vectors of the merchant words in the first text are added and averaged to obtain the vector information of the merchant to be calibrated.
[0139] C7. The electronic device according to C4, wherein the step of inputting the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine comprises:
[0140] Determining the cuisine vocabulary corresponding to the text information of the cuisine;
[0141] Determine a model according to the cuisine vocabulary and the vector, and obtain the vector of the cuisine vocabulary;
[0142] The vector of the cuisine vocabulary is used as the vector information of the cuisine.
[0143] C8. The electronic device according to C1, wherein the calculating the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine and obtaining a distance set comprises:
[0144] Calculate the cosine similarity between the vector information of the merchant to be calibrated and the vector information of each cuisine;
[0145] Obtaining the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines according to the cosine similarity;
[0146] The distance set is obtained according to the distance.
[0147] C9. According to the electronic device described in any one of C5 to C8, the description information of the merchant to be calibrated includes: the name of the merchant to be calibrated, the product name of the merchant to be calibrated, and the user evaluation of the merchant to be calibrated.
[0148] The embodiment of the present application discloses D1. a non-volatile storage medium for storing a computer-readable program, wherein the computer-readable program is used for a computer to execute a cuisine calibration method as described in any one of A1 to A9.
Claims
1. A method for calibrating a cuisine. It is characterized in that Used in electronic equipment, including: Inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the vocabulary features of the text information of the merchant to be calibrated, and the vector determination model contains the corresponding relationship between the vocabulary and the storage location identifier of the vocabulary vector; the text information of the merchant to be calibrated is obtained by sequentially connecting the word segmentation result and the single character segmentation result of the description information of the merchant to be calibrated; The step of inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated includes: adding and averaging the vectors of each merchant word in the text information of the merchant to be calibrated to obtain the vector information of the merchant to be calibrated; Inputting the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the vocabulary features of the text information of the cuisine, and the number of the cuisines is at least two; Calculating the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines and obtaining a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines; The cuisine of the merchant to be calibrated is calibrated according to the distance set.
2. The cuisine calibration method according to claim 1, It is characterized in that The step of calibrating the cuisine of the merchant to be calibrated according to the distance set includes: Determine at least one cuisine corresponding to a distance in the distance set that is less than a preset threshold; The merchant to be calibrated is calibrated according to the at least one cuisine.
3. The cuisine calibration method according to claim 1, It is characterized in that Before inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, the method further includes: Obtaining description information of sample merchants, wherein the number of the sample merchants is at least two; Obtaining sample text information according to the description information of the sample merchant, wherein the sample text information includes at least two sample words; Calculate the vector of each sample word in the sample text information according to the glove algorithm; The vector determination model is obtained by training the sample words and the vectors of the sample words in the sample text information, wherein the vector determination model includes a corresponding relationship between storage location identifiers for representing words and word vectors.
4. The cuisine calibration method according to claim 3, It is characterized in that The obtaining of sample text information according to the description information of the sample merchant includes: Performing word segmentation processing on the description information of the sample merchant to obtain a sample word segmentation result; Performing single-character segmentation processing on the description information of the sample merchant to obtain a sample segmentation result; The sample word segmentation result and the sample segmentation result are sequentially connected to obtain the sample text information.
5. The cuisine calibration method according to claim 4, It is characterized in that The step of inputting the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine includes: Determining the cuisine vocabulary corresponding to the text information of the cuisine; Determine a model according to the cuisine vocabulary and the vector, and obtain the vector of the cuisine vocabulary; The vector of the cuisine vocabulary is used as the vector information of the cuisine.
6. The cuisine calibration method according to claim 1, It is characterized in that The calculating the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine and obtaining a distance set includes: Calculate the cosine similarity between the vector information of the merchant to be calibrated and the vector information of each cuisine; Obtaining the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines according to the cosine similarity; The distance set is obtained according to the distance.
7. The cuisine calibration method according to any one of claims 4 to 6, It is characterized in that The description information of the merchant to be marked includes: the name of the merchant to be marked, the product name of the merchant to be marked, and the user evaluation of the merchant to be marked.
8. A cuisine calibration device, It is characterized in that include: A first determination module is used to input the text information of the merchant to be calibrated into a vector determination model to obtain the vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the vocabulary features of the text information of the merchant to be calibrated, and the vector determination model contains the corresponding relationship between the vocabulary and the storage location identifier of the vocabulary vector; the text information of the merchant to be calibrated is obtained by sequentially connecting the word segmentation result and the single character segmentation result of the description information of the merchant to be calibrated; The first determination module is specifically used to add and average the vectors of each merchant word in the text information of the merchant to be calibrated to obtain the vector information of the merchant to be calibrated; A second determination module is used to input the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the vocabulary features of the text information of the cuisine, and the number of the cuisines is at least two; A distance calculation module, used to calculate the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines and obtain a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines; The cuisine calibration module is used to calibrate the cuisine of the merchant to be calibrated according to the distance set.
9. An electronic device, It is characterized in that It includes a memory and a processor. The memory stores computer programs. When the processor runs the program, it executes: Inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, wherein the vector information of the merchant to be calibrated represents the vocabulary features of the text information of the merchant to be calibrated, and the vector determination model contains the corresponding relationship between the vocabulary and the storage location identifier of the vocabulary vector; the text information of the merchant to be calibrated is obtained by sequentially connecting the word segmentation result and the single character segmentation result of the description information of the merchant to be calibrated; The step of inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated includes: adding and averaging the vectors of each merchant word in the text information of the merchant to be calibrated to obtain the vector information of the merchant to be calibrated; Inputting the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine, wherein the vector information of the cuisine represents the vocabulary features of the text information of the cuisine, and the number of the cuisines is at least two; Calculating the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines and obtaining a distance set, wherein the distance set includes the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines; The cuisine of the merchant to be calibrated is calibrated according to the distance set.
10. The electronic device according to claim 9, It is characterized in that The step of calibrating the cuisine of the merchant to be calibrated according to the distance set includes: Determine at least one cuisine corresponding to a distance in the distance set that is less than a preset threshold; The merchant to be calibrated is calibrated according to the at least one cuisine.
11. The electronic device according to claim 9, It is characterized in that Before inputting the text information of the merchant to be calibrated into the vector determination model to obtain the vector information of the merchant to be calibrated, the method further includes: Obtaining description information of sample merchants, wherein the number of the sample merchants is at least two; Obtaining sample text information according to the description information of the sample merchant, wherein the sample text information includes at least two sample words; Calculate the vector of each sample word in the sample text information according to the glove algorithm; The vector determination model is obtained by training the sample words and the vectors of the sample words in the sample text information, wherein the vector determination model includes a corresponding relationship between storage location identifiers for representing words and word vectors.
12. The electronic device according to claim 11, It is characterized in that The obtaining of sample text information according to the description information of the sample merchant includes: Performing word segmentation processing on the description information of the sample merchant to obtain a sample word segmentation result; Performing single-character segmentation processing on the description information of the sample merchant to obtain a sample segmentation result; The sample word segmentation result and the sample segmentation result are sequentially connected to obtain the sample text information.
13. The electronic device according to claim 12, It is characterized in that The step of inputting the text information of the cuisine into the vector determination model to obtain the vector information of the cuisine includes: Determining the cuisine vocabulary corresponding to the text information of the cuisine; Determine a model according to the cuisine vocabulary and the vector, and obtain the vector of the cuisine vocabulary; The vector of the cuisine vocabulary is used as the vector information of the cuisine.
14. The electronic device according to claim 9, It is characterized in that The calculating the distance between the vector information of the merchant to be calibrated and the vector information of each cuisine and obtaining a distance set includes: Calculate the cosine similarity between the vector information of the merchant to be calibrated and the vector information of each cuisine; Obtaining the distance between the vector information of the merchant to be calibrated and the vector information of each of the cuisines according to the cosine similarity; The distance set is obtained according to the distance.
15. The electronic device according to any one of claims 12 to 14, It is characterized in that The description information of the merchant to be marked includes: the name of the merchant to be marked, the product name of the merchant to be marked, and the user evaluation of the merchant to be marked.
16. A non-volatile storage medium for storing a computer-readable program, It is characterized in that The computer-readable program is used for enabling a computer to execute the cuisine calibration method according to any one of claims 1 to 7.
