Enterprise profit prediction method and system

Abstract

The invention provides an enterprise profit prediction method and system, and the method comprises the steps: obtaining a monthly profit data set comprising a plurality of monthly profit data and a monthly evaluation data set comprising a plurality of monthly evaluation data, the monthly evaluation data comprising a plurality of user IDs, direct scoring data and comment data; acquiring a first score based on the direct scoring data, acquiring a second initial score based on the comment data, acquiring a second final score based on the first score and the second initial score, and further initializing a total score; optimizing the initial total score value into a final total score value based on the reputation score so as to obtain a monthly score value; and training the initial prediction neural network model based on the monthly score value and the monthly profit data to obtain a final prediction neural network model for profit prediction. By means of the method, more basic data are introduced on the basis of reading the profit data set, the purpose of enriching the predicted basic data volume is achieved, and the profit prediction accuracy of the type of enterprises is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data prediction, and particularly relates to a method and system for predicting corporate profits. Background Art

[0002] The progress of the times is not only reflected in the development of technology, but also in the openness of information. Financial indicator data can reflect the financial status and operating results of an enterprise during a certain period to a certain extent, and the prediction of corporate profits can also be achieved based on multi-period financial indicator data.

[0003] For listed enterprises with a long establishment time, they usually have financial indicator data with a long duration. By comprehensively analyzing the characteristics of the financial indicator data, the prediction of corporate profits can be completed more quickly and accurately.

[0004] However, for small and medium-sized catering enterprises with a short establishment time, the cycle of their financial indicator data is short. Affected by the small amount of data, the method of predicting corporate profits based on financial indicator data is prone to deviation in the prediction of corporate profits. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a method and system for predicting corporate profits, aiming to solve the technical problem that in the prior art, for small and medium-sized catering enterprises with a short establishment time, the method of predicting corporate profits based on financial indicator data is affected by the small amount of data and is prone to deviation in the prediction of profits.

[0006] To achieve the above purpose, in the first aspect, an embodiment of the present application provides a method for predicting corporate profits, including the following steps:

[0007] Obtain a monthly profit dataset and a monthly evaluation dataset corresponding to small and medium-sized catering enterprises. The monthly profit dataset includes several monthly profit data, and the monthly evaluation dataset includes several monthly evaluation data. The monthly evaluation data includes several user IDs, direct scoring data corresponding to the user IDs, and comment data;

[0008] Obtain a first score based on the direct scoring data, and obtain a second initial score based on the comment data. Perform a mapping process on the second initial score through the first score to obtain a second final score, and obtain an initial total score value corresponding to the user ID through the first score and the second final score;

[0009] Obtain the credit score corresponding to the user ID, optimize the initial total score value based on the credit score to obtain the final total score value, and obtain the monthly score value based on several of the final total score values, where the monthly score value corresponds to the monthly profit data;

[0010] Construct an initial prediction neural network model, train the initial prediction neural network model based on the monthly score value and the monthly profit data to obtain the final prediction neural network model, and perform profit prediction based on the final prediction neural network model.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: when the amount of data of the monthly profit data of small and medium-sized catering enterprises with a short establishment time is small, the impact of user evaluations on such enterprises' profit situations is considered. When performing profit prediction, the concept of monthly evaluation data is introduced, the first score value and the second final score value are obtained based on the monthly evaluation data, and further the final total score value is obtained. Profit prediction is carried out in a manner combining the monthly profit data and the final total score value data, that is, more basic data is introduced through impact factors related to profit, achieving the purpose of enriching the amount of basic data for prediction and improving the accuracy of profit prediction for such enterprises.

[0012] Furthermore, the direct score data includes comprehensive score, taste score, service score, and environment score, and the formula for obtaining the first score value is:

[0013]

[0014] Wherein, F 1 represents the first score value, Z p represents the comprehensive score, K p represents the taste score, F p represents the service score, H p represents the environment score.

[0015] Even further, the step of obtaining the second initial score value based on the comment data includes:

[0016] Perform word segmentation on the comment data to obtain several comment words;

[0017] Compare the comment words with a preset dictionary to separate several of the comment words into positive words, negative words, and irrelevant words;

[0018] Obtain the second initial score value based on the number of the positive words and the negative words.

[0019] Even further, the formula for obtaining the second initial score value is:

[0020] Fc2 = max((1 * n z - 1 * n f ), 0),

[0021] where Fc 2 represents the second initial score, n z represents the number of positive words, and n f represents the number of negative words. max(·) represents the maximum value function.

[0022] Furthermore, the step of mapping the second initial score through the first score to obtain the second final score includes:

[0023] Obtain the first maximum score and the first minimum score of the first score, and obtain the second maximum score and the second minimum score of the second initial score;

[0024] Obtain the first score range between the first maximum score and the first minimum score, and obtain the second score range between the second maximum score and the second minimum score;

[0025] Map the second maximum score to the first maximum score and map the second minimum score to the first minimum score to obtain the mapping relationship between the first score range and the second score range;

[0026] Adjust the second initial score to the second final score based on the mapping relationship.

[0027] Furthermore, the calculation formula for the initial total score is:

[0028]

[0029] where ZP i represents the initial total score corresponding to the i-th user ID, and both α and β represent score weights, represents the first score corresponding to the i-th user ID, represents the second final score corresponding to the i-th user ID.

[0030] Furthermore, the step of optimizing the initial total score value based on the credit score to obtain the final total score value includes:

[0031] Set a score region corresponding to the credit score, and the score region includes several score segments and adjustment ratios corresponding to the score segments;

[0032] Set a score value region corresponding to the initial total score value, and the score value region includes a high score segment, a medium score segment, and a low score segment;

[0033] Compare the credit score with the score range to select a final adjustment ratio from several adjustment ratios, and compare the initial total score value with the score range;

[0034] If the initial total score value is in the high score segment, lower the initial total score value by the final adjustment ratio to obtain the final total score value; if the initial total score value is in the low score segment, raise the initial total score value by the final adjustment ratio to obtain the final total score value; if the initial total score value is in the medium score segment, determine the initial total score value as the final total score value.

[0035] Furthermore, the step of training the initial prediction neural network model based on the monthly score value and the monthly profit data includes:

[0036] Partition the monthly profit data into pre-training monthly data and current training month data based on time sequence, and select the monthly score value corresponding to the pre-training monthly data as the pre-training monthly score;

[0037] Use the pre-training monthly data and the pre-training monthly score as the input values of the initial prediction neural network model, and use the current training month data as the output value of the initial prediction neural network model to train the initial prediction neural network model.

[0038] In a second aspect, an embodiment of the present application provides an enterprise profit prediction system, which is applied to the enterprise profit prediction method described in the first aspect above. The system includes:

[0039] An acquisition module, configured to acquire a monthly profit data set and a monthly evaluation data set corresponding to small and medium-sized catering enterprises. The monthly profit data set includes several monthly profit data, and the monthly evaluation data set includes several monthly evaluation data. The monthly evaluation data includes several user IDs, direct score data corresponding to the user IDs, and comment data;

[0040] An evaluation module, configured to obtain a first score based on the direct score data, obtain a second initial score based on the comment data, perform mapping processing on the second initial score through the first score to obtain a second final score, and obtain an initial total score value corresponding to the user ID through the first score and the second final score;

[0041] An adjustment module, configured to obtain a credit score corresponding to the user ID, perform optimization processing on the initial total score value based on the credit score to obtain a final total score value, and obtain a monthly score value based on several final total score values. The monthly score value corresponds to the monthly profit data;

[0042] An execution module, configured to build an initial prediction neural network model, train the initial prediction neural network model based on the monthly score value and the monthly profit data to obtain a final prediction neural network model, and perform profit prediction based on the final prediction neural network model.

[0043] In a third aspect, an embodiment of the present application provides a computer, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the enterprise profit prediction method as described in the first aspect above is implemented.

[0044] In a fourth aspect, an embodiment of the present application provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the enterprise profit prediction method as described in the first aspect above is implemented. Description of the Drawings

[0045] Figure 1 It is a flowchart of the enterprise profit prediction method in the first embodiment of the present invention;

[0046] Figure 2 It is a structural block diagram of the enterprise profit prediction system in the second embodiment of the present invention;

[0047] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. Specific Embodiments

[0048] To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present invention more thorough and comprehensive.

[0049] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly on the other element or there can also be a middle element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be a middle element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are only for the purpose of illustration.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used in the description of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0051] Please refer to Figure 1 , the enterprise profit prediction method provided by the first embodiment of the present invention includes the following steps:

[0052] S10: Obtain a monthly profit data set and a monthly evaluation data set corresponding to small and medium-sized catering enterprises. The monthly profit data set includes several months of profit data, and the monthly evaluation data set includes several months of evaluation data. The monthly evaluation data includes several user IDs, direct score data corresponding to the user IDs, and comment data;

[0053] In some embodiments, profit prediction can also be performed on other marketing enterprises that exist on the network platform and have marketing products, and users can evaluate and comment on them on the network platform. In this embodiment, the direct score data includes a comprehensive score, a taste score, a service score, and an environment score. The scores of the comprehensive score, the taste score, the service score, and the environment score are all 0-5 points. Specifically, the link URL can be obtained according to the merchant ID corresponding to the small and medium-sized catering enterprise, the comment page can be parsed from the link URL, and the monthly evaluation data can be extracted from the comment page.

[0054] S20: Obtain a first score based on the direct score data, and obtain a second initial score based on the comment data. Perform mapping processing on the second initial score through the first score to obtain a second final score, and obtain an initial total score value corresponding to the user ID through the first score and the second final score;

[0055] The formula for obtaining the first score is:

[0056]

[0057] Wherein, F 1 represents the first score, Z p represents the comprehensive score, K p represents the taste score, F p represents the service score, and H p represents the environment score.

[0058] The step S20 includes:

[0059] S210: Perform word segmentation processing on the comment data to obtain several comment words;

[0060] Word segmentation processing can now be processed by many algorithms, and will not be elaborated here.

[0061] S220: Compare the comment words with a preset dictionary to separate several of the comment words into positive words, negative words, and irrelevant words;

[0062] The preset dictionary includes a positive dictionary and a negative dictionary. The comment words are respectively matched with the positive dictionary and the negative dictionary to select the comment words as the positive words or the negative words according to the matching results. If the comment words do not match both the positive dictionary and the negative dictionary, they are selected as the irrelevant words, and the irrelevant words are proposed.

[0063] S230: Obtain the second initial score based on the quantities of the positive words and the negative words;

[0064] The formula for obtaining the second initial score is:

[0065] Fc 2 = max((1 * n z - 1 * n f ), 0),

[0066] where Fc 2 represents the second initial score, n z represents the quantity of positive words, n f represents the quantity of negative words, and max(·) represents the maximum value function. It can be understood that the minimum value of the second initial score is 0 points.

[0067] S240: Obtain the first maximum score and the first minimum score of the first score, and obtain the second maximum score and the second minimum score of the second initial score;

[0068] S250: Obtain the first score range between the first maximum score and the first minimum score, and obtain the second score range between the second maximum score and the second minimum score;

[0069] S260: Map the second maximum score to the first maximum score and map the second minimum score to the first minimum score to obtain the mapping relationship between the first score range and the second score range;

[0070] S270: Adjust the second initial score to the second final score based on the mapping relationship;

[0071] Based on the description in step S10, in this embodiment, the first maximum score is 0 points and the first minimum score is 5 points. If the second maximum score is 30 points and the second minimum score is 5 points, then the first score range is 0 points to 5 points, and the second score range is 5 points to 30 points. If the second initial score is 5 points, then the second final score is 0 points. If the second initial score is 30 points, then the second final score is 5 points. Based on this mapping relationship, if the second initial score is 15 points, then the second final score is 2 points.

[0072] The calculation formula of the initial total score is as follows:

[0073]

[0074] where ZP i represents the initial total score corresponding to the i-th user ID, and both α and β represent score weights. represents the first score corresponding to the i-th user ID. represents the second final score corresponding to the i-th user ID. In this embodiment, the values of α and β are both 0.5.

[0075] S30: Obtain the credit score corresponding to the user ID, optimize the initial total score value based on the credit score to obtain the final total score value, and obtain the monthly score value based on several final total score values. The monthly score value corresponds to the monthly profit data.

[0076] The credit score is the score given by the network platform for the past behaviors of the user ID. The past behaviors include whether there is malicious brushing of reviews, whether there is malicious order cancellation, etc.

[0077] The step S30 includes:

[0078] S310: Set a score area corresponding to the credit score. The score area includes several score segments and adjustment ratios corresponding to the score segments.

[0079] In this embodiment, the score area includes 5 score segments, which are 0 - 20 points, 21 - 40 points, 41 - 60 points, 61 - 80 points, and 81 - 100 points respectively. The adjustment ratios corresponding to the 5 score segments are: 80%, 60%, 40%, 20%, 0%.

[0080] S320: Set a score value area corresponding to the initial total score value. The score value area includes a high score segment, a medium score segment, and a low score segment.

[0081] In this embodiment, the high score segment is 4 - 5 points, the medium score segment is 2 - 3 points, and the low score segment is 0 - 1 point.

[0082] S330: Compare the credit score with the score area to select the final adjustment ratio from several adjustment ratios, and compare the initial total score value with the score value area.

[0083] S340: If the initial total score value is within the high score range, lower the initial total score value by the final adjustment ratio to obtain the final total score value; if the initial total score value is within the low score range, raise the initial total score value by the final adjustment ratio to obtain the final total score value; if the initial total score value is within the medium score range, determine the initial total score value as the final total score value.

[0084] By considering the credit score corresponding to the user ID and performing the optimization process on the initial total score based on the credit score, the deviation between the monthly score value and the actual situation caused by malicious evaluation can be avoided to a certain extent, improving the accuracy of subsequent profit prediction.

[0085] S40: Construct an initial prediction neural network model, train the initial prediction neural network model based on the monthly score value and the monthly profit data to obtain a final prediction neural network model, and perform profit prediction based on the final prediction neural network model;

[0086] The application of neural network models is extensive and will not be elaborated here. In this embodiment, the initial prediction neural network model is an LSTM neural network model.

[0087] Specifically, the step S40 includes:

[0088] S410: Partition the monthly profit data into pre-training monthly data and current training month data based on the time sequence, and select the monthly score value corresponding to the pre-training monthly data as the pre-training monthly score;

[0089] S420: Use the pre-training monthly data and the pre-training monthly score as the input values of the initial prediction neural network model, and use the current training month data as the output value of the initial prediction neural network model to train the initial prediction neural network model;

[0090] It should be noted that several sets of the monthly profit data and several sets of the monthly score values can be combined into multiple combinations of ((pre-training monthly data, pre-training monthly score) - current training month data), and then through repeated training multiple times, the prediction accuracy of the initial prediction neural network model can be improved to form the final prediction neural network model.

[0091] In the case where the amount of data of the monthly profit data of small and medium-sized catering enterprises with a short establishment time is small, the impact of user evaluations of such enterprises on their profit situations is considered. When making profit predictions, the concept of the monthly evaluation data is introduced. Based on the monthly evaluation data, the first score and the second final score are obtained, and further the final total score value is obtained. Profit predictions are made in a way that combines the monthly profit data and the final total score value data, that is, more basic data is introduced through the impact factors related to profit, achieving the purpose of enriching the amount of basic data for prediction and improving the accuracy of profit predictions for such enterprises.

[0092] Please refer to Figure 2 , the second embodiment of the present invention provides an enterprise profit prediction system. This system is applied to the enterprise profit prediction method in the above-mentioned embodiment, and those that have been described will not be repeated. As used hereinafter, terms such as "module", "unit", "sub-unit", etc. can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0093] The system includes:

[0094] An acquisition module 10, configured to acquire a monthly profit data set and a monthly evaluation data set corresponding to small and medium-sized catering enterprises. The monthly profit data set includes a number of monthly profit data, and the monthly evaluation data set includes a number of monthly evaluation data. The monthly evaluation data includes a number of user IDs, direct score data corresponding to the user IDs, and comment data;

[0095] An evaluation module 20, configured to obtain a first score based on the direct score data, obtain a second initial score based on the comment data, perform a mapping process on the second initial score through the first score to obtain a second final score, and obtain an initial total score value corresponding to the user ID through the first score and the second final score;

[0096] The acquisition module 20 includes:

[0097] A first unit, configured to perform word segmentation on the comment data to obtain a number of comment words;

[0098] A second unit, configured to compare the comment words with a preset dictionary to separate a number of the comment words into positive words, negative words, and irrelevant words;

[0099] A third unit, configured to obtain the second initial score based on the number of the positive words and the negative words;

[0100] A fourth unit for obtaining the first maximum score and the first minimum score of the first score value, and obtaining the second maximum score and the second minimum score of the second initial score value;

[0101] A fifth unit for obtaining a first score range between the first maximum score and the first minimum score, and obtaining a second score range between the second maximum score and the second minimum score;

[0102] A sixth unit for mapping the second maximum score to the first maximum score and mapping the second minimum score to the first minimum score to obtain a mapping relationship between the first score range and the second score range;

[0103] A seventh unit for adjusting the second initial score value to the second final score value based on the mapping relationship;

[0104] An adjustment module 30 for obtaining a credit score corresponding to the user ID, optimizing the initial total score value based on the credit score to obtain a final total score value, and obtaining a monthly score value based on a plurality of the final total score values, where the monthly score value corresponds to the monthly profit data;

[0105] The adjustment module 30 includes:

[0106] An eighth unit for setting a score region corresponding to the credit score, where the score region includes a plurality of score segments and adjustment ratios corresponding to the score segments;

[0107] A ninth unit for setting a score value region corresponding to the initial total score value, where the score value region includes a high score segment, a medium score segment, and a low score segment;

[0108] A tenth unit for comparing the credit score with the score region to select a final adjustment ratio from a plurality of the adjustment ratios, and comparing the initial total score value with the score value region;

[0109] An eleventh unit for, if the initial total score value is in the high score segment, reducing the initial total score value by the final adjustment ratio to obtain a final total score value; if the initial total score value is in the low score segment, increasing the initial total score value by the final adjustment ratio to obtain a final total score value; if the initial total score value is in the medium score segment, determining the initial total score value as the final total score value;

[0110] Execution module 40, configured to build an initial prediction neural network model, train the initial prediction neural network model based on the monthly score value and the monthly profit data to obtain a final prediction neural network model, and perform profit prediction based on the final prediction neural network model;

[0111] The execution module 40 includes:

[0112] The twelfth unit is configured to partition the monthly profit data into pre-training monthly data and current training monthly data based on time sequence, and select the monthly score value corresponding to the pre-training monthly data as the pre-training monthly score;

[0113] The thirteenth unit is configured to use the pre-training monthly data and the pre-training monthly score as input values of the initial prediction neural network model, and use the current training monthly data as the output value of the initial prediction neural network model to train the initial prediction neural network model.

[0114] The present invention also provides a computer, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the enterprise profit prediction method described in the above technical solution is implemented.

[0115] The present invention also provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the enterprise profit prediction method described in the above technical solution is implemented.

[0116] In the description of this specification, the descriptions with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0117] The above embodiments only represent several implementation manners of the present invention. The descriptions are relatively specific and detailed, but should not be construed as a limitation on the scope of the invention patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several deformations and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the invention patent of the present invention should be subject to the appended claims.

Claims

1. A method for predicting corporate profits, characterized in that: The following steps are involved: Acquire a monthly profit data set and a monthly evaluation data set corresponding to small and medium-sized catering enterprises, wherein the monthly profit data set includes profit data for several months, the monthly evaluation data set includes evaluation data for several months, and the monthly evaluation data includes several user IDs, direct rating data corresponding to the user IDs, and comment data; Obtaining a first score based on the direct scoring data, and obtaining a second initial score based on the comment data, mapping the second initial score through the first score to obtain a second final score, and obtaining an initial total score corresponding to the user ID through the first score and the second final score; Obtaining a reputation score corresponding to the user ID, optimizing the initial total score value based on the reputation score to obtain a final total score value, and obtaining a monthly score value based on a plurality of the final total score values, wherein the monthly score value corresponds to the monthly profit data; An initial prediction neural network model is constructed, and the initial prediction neural network model is trained based on the monthly score value and the monthly profit data to obtain a final prediction neural network model, and profit prediction is performed based on the final prediction neural network model.

2. The enterprise profit forecasting method according to claim 1, characterized in that: The direct scoring data includes comprehensive score, taste score, service score and environment score. The formula for obtaining the first score is: Among them, F1 represents the first score, Z p represents the comprehensive score, K p Indicates the taste score, F p represents the service rating, H p Represents the environmental score.

3. The enterprise profit forecasting method according to claim 1, characterized in that: The step of obtaining a second initial score based on the comment data comprises: Performing word segmentation processing on the comment data to obtain a number of comment words; Comparing the review words with a preset dictionary to separate a number of the review words into positive words, negative words and irrelevant words; The second initial score is obtained based on the number of the positive words and the negative words.

4. The enterprise profit forecasting method according to claim 3, characterized in that: The formula for obtaining the second initial score is: Fc2 = max((1*n z -1*n f ),0), Wherein, Fc2 represents the second initial score, n z Indicates the number of positive words, n f represents the number of negative words, and max(·) represents the maximum value function.

5. The enterprise profit forecasting method according to claim 1, characterized in that: The step of mapping the second initial score by the first score to obtain the second final score comprises: Obtaining a first maximum score and a first minimum score of the first score, and obtaining a second maximum score and a second minimum score of the second initial score; Obtaining a first score interval between the first maximum score and the first minimum score, and obtaining a second score interval between the second maximum score and the second minimum score; Mapping the second maximum score to the first maximum score, and mapping the second minimum score to the first minimum score, to obtain a mapping relationship between the first score interval and the second score interval; The second initial score is adjusted to the second final score based on the mapping relationship.

6. The enterprise profit forecasting method according to claim 1, characterized in that: The calculation formula of the initial total score is: Among them, ZP i represents the initial total score corresponding to the i-th user ID, α and β both represent the score weights, represents the first score corresponding to the i-th user ID, represents the second final score corresponding to the i-th user ID.

7. The enterprise profit forecasting method according to claim 1, characterized in that: The step of optimizing the initial total score value based on the reputation score to obtain a final total score value comprises: Setting a scoring area corresponding to the reputation score, the scoring area including a plurality of scoring segments and adjustment ratios corresponding to the scoring segments; Setting a score area corresponding to the initial total score, the score area including a high score segment, a medium score segment and a low score segment; Comparing the reputation score with the score region to select a final adjustment ratio from a plurality of adjustment ratios, and comparing the initial total score with the score region; If the initial total score value is in the high score segment, the initial total score value is adjusted down by the final adjustment ratio to obtain the final total score value; if the initial total score value is in the low score segment, the initial total score value is adjusted up by the final adjustment ratio to obtain the final total score value; if the initial total score value is in the medium score segment, the initial total score value is determined as the final total score value.

8. The enterprise profit forecasting method according to claim 1, characterized in that: The step of training the initial prediction neural network model based on the monthly rating value and the monthly profit data comprises: Separating the monthly profit data into pre-training month data and training month data based on the time sequence, and selecting the monthly score value corresponding to the pre-training month data as the pre-training month score; The initial prediction neural network model is trained by using the data from the month before training and the score from the month before training as input values ​​of the initial prediction neural network model, and using the data from the current month of training as output values ​​of the initial prediction neural network model.

9. A corporate profit forecasting system, applied to the corporate profit forecasting method according to any one of claims 1 to 8, characterized in that: The system comprises: An acquisition module, used to acquire a monthly profit data set and a monthly evaluation data set corresponding to small and medium-sized catering enterprises, wherein the monthly profit data set includes profit data of several months, the monthly evaluation data set includes evaluation data of several months, and the monthly evaluation data includes several user IDs, direct rating data corresponding to the user IDs, and comment data; An evaluation module, configured to obtain a first score based on the direct scoring data, and obtain a second initial score based on the comment data, map the second initial score with the first score to obtain a second final score, and obtain an initial total score corresponding to the user ID with the first score and the second final score; an adjustment module, configured to obtain a reputation score corresponding to the user ID, optimize the initial total score value based on the reputation score to obtain a final total score value, and obtain a monthly score value based on a plurality of the final total score values, wherein the monthly score value corresponds to the monthly profit data; An execution module is used to construct an initial prediction neural network model, train the initial prediction neural network model based on the monthly score value and the monthly profit data to obtain a final prediction neural network model, and perform profit prediction based on the final prediction neural network model.

10. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the enterprise profit forecasting method as described in the first aspect above is implemented.

Images