Supply and demand matching method and device, equipment, computer storage medium and program product

Abstract

The invention discloses a supply and demand matching method, device and equipment, a computer storage medium and a program product, and is applied to the technical field of supply and demand matching, and the method comprises the steps: obtaining demand information including demand description, release time and position information, and multiple pieces of supply information including supply description, release time, position information and historical evaluation data; calculating an initial matching degree between the demand information and each piece of supply information according to the demand description and the supply description; calculating a time attenuation factor of the demand information and each piece of supply information according to the demand release time, each piece of supply release time and each piece of historical evaluation data; calculating spatial attenuation factors of the demand information and each piece of supply information according to the demand position information and the supply position information; calculating a target matching degree between the demand information and each piece of supply information according to the time attenuation factor, the space attenuation factor and the initial matching degree; and determining at least one piece of supply information matched with the demand information according to the target matching degree. The supply and demand matching accuracy can be improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of supply and demand matching, and particularly relates to a supply and demand matching method, device, equipment, computer storage medium and program product. BACKGROUND

[0002] In various online matching platforms, such as investment and financing platforms, commodity procurement platforms and service docking platforms, the core function is to efficiently match the supply information and demand information published by users.

[0003] At present, supply and demand matching mainly depends on content similarity calculation matching degree, such as matching through text similarity of product name and category. However, based on content similarity calculation matching degree, the value decay of supply and demand information over time cannot be perceived, and in time-sensitive scenarios such as investment and financing and time-limited services, the matching accuracy significantly decreases, resulting in a large number of seemingly matched information being recommended to users, which does not meet the actual needs of users.

[0004] Therefore, the prior art has the problems of low matching accuracy and recommended results not meeting the actual needs of users. SUMMARY

[0005] The embodiments of the application provide a supply and demand matching method, device, equipment, computer storage medium and program product, which can improve the matching accuracy, ensure that the matching result meets the actual needs, reduce the user screening cost and promote the transaction to be reached.

[0006] In a first aspect, the embodiments of the application provide a supply and demand matching method, which comprises: obtaining demand information and a plurality of supply information, wherein the demand information comprises demand description, demand publishing time and demand location information, and the supply information comprises supply description, supply publishing time, supply location information and historical evaluation data; calculating an initial matching degree of the demand information and each of the supply information according to the demand description and the supply description; calculating a time decay factor of the demand information and each of the supply information according to the demand publishing time, each of the supply publishing time and each of the historical evaluation data; calculating a spatial decay factor of the demand information and each of the supply information according to the demand location information and the supply location information; calculating a target matching degree of the demand information and each of the supply information according to the time decay factor, the spatial decay factor and the initial matching degree; determining at least one of the supply information matched to the demand information according to the target matching degree.

[0007] In some possible implementation manners, the calculating the time decay factor of the demand information and each of the supply information according to the demand publishing time, each of the supply publishing time and each of the historical evaluation data comprises: calculating a time decay coefficient of each of the supply information according to each of the historical evaluation data; calculating a time decay factor corresponding to each of the supply information according to each of the time decay coefficient and a time difference between the supply publishing time and the demand publishing time, the time decay factor being negatively correlated with the time difference.

[0008] In some possible implementation manners, the calculating the time decay coefficient of each of the supply information according to each of the historical evaluation data comprises: determining a quantity of positive feedback and a quantity of negative feedback in each of the historical evaluation data; calculating a quality score of each of the supply information based on the quantity of positive feedback and the quantity of negative feedback; determining the time decay coefficient corresponding to the supply information according to a plurality of first preset mapping relationships, wherein the time decay coefficient is positively correlated with the quality score.

[0009] In some possible implementation manners, the calculating the time decay factor corresponding to each of the supply information according to each of the time decay coefficient and the time difference between the supply publishing time and the demand publishing time comprises: calculating the time decay factor by using the following formula: α = e^(-λ*t), wherein α is the time decay factor, λ is the time decay coefficient, and t is the time difference between the supply publishing time and the demand publishing time.

[0010] In some possible implementation manners, the demand position information comprises at least one demand position, the supply position information comprises at least one supply position, and the calculating the spatial decay factor of the demand information and each of the supply information according to the demand position information and the supply position information comprises: for each of the demand positions, determining a nearest supply position with a minimum spatial distance to the demand position in each of the supply information; calculating a comprehensive spatial distance of the demand information and each of the supply information, the comprehensive spatial distance being a sum of spatial distances between each of the demand positions and the nearest supply position in the supply information; calculating the spatial decay factor according to each of the comprehensive spatial distance.

[0011] In some possible implementation manners, the calculating the spatial decay factor according to each of the comprehensive spatial distance comprises: The spatial decay factor is calculated by the following formula: β = 1 / (1+d), wherein β is the spatial decay factor, and d is the comprehensive spatial distance.

[0012] In a second aspect, the embodiments of the present application provide a supply-demand matching device, which comprises: a obtaining module, configured to obtain demand information and a plurality of supply information, wherein the demand information comprises demand description, demand publishing time and demand location information, and the supply information comprises supply description, supply publishing time, supply location information and historical evaluation data; a first calculating module, configured to calculate initial matching degrees of the demand information and each of the supply information according to the demand description and the supply description; a second calculating module, configured to calculate time decay factors of the demand information and each of the supply information according to the demand publishing time, each of the supply publishing time and each of the historical evaluation data; a third calculating module, configured to calculate spatial decay factors of the demand information and each of the supply information according to the demand location information and the supply location information; a fourth calculating module, configured to weight each of the initial matching degrees by using the time decay factors and the spatial decay factors, and obtain target matching degrees of the demand information and each of the supply information; a determining module, configured to determine at least one of the supply information matched to the demand information according to the target matching degrees.

[0013] In a third aspect, the embodiments of the present application provide an electronic device, which comprises: a processor and a memory storing computer program instructions; the processor executes the computer program instructions to implement the supply-demand matching method of any one of the above.

[0014] In a fourth aspect, the embodiments of the present application provide a computer storage medium, which stores computer program instructions, and the computer program instructions are executed by a processor to implement the supply-demand matching method of any one of the above.

[0015] In a fifth aspect, the embodiments of the present application provide a computer program product, and instructions in the computer program product are executed by a processor of an electronic device to enable the electronic device to execute the supply-demand matching method of any one of the above.

[0016] The supply-demand matching method, device, equipment, computer storage medium and computer program product provided by the embodiments of the present application can make the time sensitivity of different supply information different by calculating the time decay factor according to the demand publishing time, the supply publishing time and the historical evaluation data, so that the high evaluation supply can still maintain a high matching degree as time elapses. The spatial decay factor is calculated according to the demand location information and the supply location information, so as to reflect the influence of spatial distance and region on the matching result, improve the matching accuracy, ensure that the matching result meets the actual demand, reduce the user screening cost, and promote the transaction to be reached. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly introduced. Those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0018] Figure 1 is a flowchart of a supply-demand matching method provided by an embodiment of the present application; Figure 2 is a structural schematic diagram of a supply-demand matching device provided by another embodiment of the present application; Figure 3 is a structural schematic diagram of an electronic device provided by another embodiment of the present application. DETAILED DESCRIPTION

[0019] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application more clear, the present application will be further described in detail below in combination with the drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, but not to limit the present application. The present application can be implemented without some of these specific details by those skilled in the art. The following description of the embodiments is only intended to provide a better understanding of the present application by showing examples of the present application.

[0020] It should be noted that the relative terms, such as first and second, and the like, are used herein only to distinguish one entity or operation from another, and do not necessarily require or imply any actual such relationship or order between or among such entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element preceded by "comprises... a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the listed element.

[0021] It should be noted that the acquisition, storage, use and processing of data in the embodiments of the present application comply with the relevant provisions of national laws and regulations.

[0022] It should be noted that in the embodiments of the present application, some software, components, models and other existing solutions in the industry may be mentioned, which should be considered as exemplary, and the purpose is only to illustrate the feasibility of the implementation of the technical solutions of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0023] First, the background technology involved in the present application is described. In various online matching platforms, such as investment and financing platforms, commodity procurement platforms, and service docking platforms, the core function is to efficiently match the supply information and demand information published by users.

[0024] At present, supply and demand matching mainly relies on content similarity calculation to match the degree, such as matching through text similarity of product name and category. However, based on content similarity calculation to match the degree cannot perceive the value decay of supply and demand information over time, and the matching accuracy significantly decreases in time-sensitive scenarios such as investment and financing and time-limited services, resulting in a large number of seemingly matched information being recommended to users, which does not meet the actual needs of users, and there is a problem of low matching accuracy and recommended results not meeting the actual needs of users.

[0025] To solve the problems in the prior art, the embodiments of the present application provide a supply and demand matching method, device, equipment, computer storage medium and program product.

[0026] First, the supply and demand matching method provided by the embodiments of the present application is introduced.

[0027] Figure 1 A flowchart of a supply and demand matching method provided by an embodiment of the present application is shown. As shown in Figure 1 The method can include steps 101 to 106.

[0028] At step 101, obtain demand information and a plurality of supply information, the demand information comprising demand description, demand publishing time and demand location information, the supply information comprising supply description, supply publishing time, supply location information and historical evaluation data.

[0029] In some embodiments, the demand information and the supply information refer to information published by users, such as investment parties, financing parties, purchasers and suppliers, in relation to demands. When the user is an investment party, the demand information is investment information and the supply information is financing information. It can be understood that in the investment and financing scenario, when the demand information is published by a financing party user, the demand information published by other investment party users serves as the supply information of the demand information published by the financing party user; when the demand information is published by an investment party user, the demand information published by other financing party users serves as the supply information of the demand information published by the investment party user. Similarly, in the goods buying and selling scenario, when the demand information is purchase information, the supply information is supply information; when the demand information is supply information, the supply information is purchase information.

[0030] The demand description and the supply description can be text fields. In the investment and financing scenario, the demand description is used to describe financing demand or investment demand. In the goods buying and selling scenario, the demand description and the supply description are used to describe the specification parameters of the goods, such as category, brand, inventory, price, etc.

[0031] Exemplarily, the demand description and the supply description can be, for example, the expected financing is used for product research and development and market expansion in the future 3-5 years, and the current round is B round; the investment period is “3-5 years” and the preference is “B round” of “artificial intelligence” enterprises; 1000 cases of mineral water are purchased, the specification is 500 ml x 24 bottles, and the brand preference is A or B; A brand mineral water is supplied, 1000 cases of current goods are supplied, the specification is 500 ml x 24 bottles, and 100 pieces of electronic equipment are purchased.

[0032] The demand publishing time and the supply publishing time can be a timestamp, which is used to record the publishing or updating time of the information. For example, in the investment and financing scenario, the time of publishing the financing demand and the time of publishing or updating the investment demand are recorded. In the goods buying and selling scenario, the time of publishing the purchase order and the time of publishing or updating the sale information are recorded.

[0033] The demand location information and the supply location information can be geographic location information data. In the investment and financing scenario, it can be the registered place of the financing party company or the target market area, or the office of the investment party or the investment area preference. In the goods buying and selling scenario, it can be the goods distribution address or the goods shipping location.

[0034] The historical evaluation data is used to reflect the quality and reliability of the past service of the supplier, and can be the score of the historical user matched to the supply information, the number of good and bad comments, the complaint rate, the post-transaction feedback, etc. In the investment and financing scene, the historical evaluation data can be the investment success rate of the investor, the return rate, the financing party evaluation, etc., or the past project success rate of the financing party, the repayment record, the investment party evaluation, etc. In the goods buying and selling scene, the historical evaluation data can be the on-time delivery rate of the supplier, the product qualification rate, the product evaluation, the on-time payment rate of the demand party, the supplier evaluation, etc.

[0035] In step 102, the initial matching degree of the demand information and each supply information is calculated according to the demand description and the supply description.

[0036] In some embodiments, the demand description and the supply description can include at least one of structured data and unstructured data. The structured data is data with a predefined model and format specification, and common fields include commodity category, service type, price, quantity, brand, financing round, investment amount range, skill level, etc. The unstructured data refers to free text without a predefined model, such as commodity description, project details, personal profile, etc.

[0037] For structured data, the first content matching degree can be calculated by, for example, rule matching and field similarity calculation. For example, the demand information is aligned with the same named fields in the supply information. For example, the "commodity category" field of the demand party is compared with the "commodity category" field of the supply party. And for different types of fields, different similarity algorithms can be used to calculate the similarity, and then the sum of the similarities of all fields is determined as the first content matching degree of the structured data. For example, for commodity category, a preset category relationship tree can be used to calculate the similarity, such as if the demand category is a smart phone and the supply category is also a smart phone, the similarity is 1.0; if the supply category is a mobile phone accessory, the similarity can be 0.3. For numerical fields such as price and amount, the relative difference or range overlap degree can be used to calculate the similarity.

[0038] For unstructured data, keyword extraction and standardization processing can be performed through natural language processing technology to convert the unstructured data into a vector. Then, the second content matching degree is calculated by the cosine similarity algorithm. Then, the initial matching degree can be calculated according to the first content matching degree and the second content matching degree, for example, the average of the first content matching degree and the second content matching degree is calculated, and the average is determined as the initial matching degree.

[0039] In some embodiments, keyword extraction and standardization can also be directly performed by natural language processing technology, the demand description is converted into a demand description vector, the supply description is converted into a supply description vector, the cosine similarity between the demand description vector and the supply description vector is calculated by a cosine similarity algorithm, and the cosine similarity is determined as the initial matching degree.

[0040] wherein the cosine similarity algorithm is wherein · denotes dot product, denotes the modulus of the vector, A and B are two vectors for which the cosine similarity needs to be calculated.

[0041] In step 103, a time decay factor of the demand information and each supply information is calculated according to the demand publishing time, each supply publishing time and historical evaluation data.

[0042] The time decay factor is used to quantify the influence of time difference on the matching degree, and the value of the time decay factor decays with the increase of time, reflecting the timeliness of the demand. The historical evaluation data is used to adjust the sensitivity of time decay, because in the investment scenario, the investment party with high historical evaluation should not be decayed too fast even if the information is published a little earlier, because their financial strength and reputation are more reliable. Similarly, in the goods buying and selling, the supplier with good reputation may be more reliable than the new supplier even if the inventory information is updated a little earlier. Therefore, the present application makes the supply information with better historical evaluation have a lower time decay factor through the historical evaluation data, avoiding that the high-quality but slightly earlier information is filtered out simply by time difference decay.

[0043] In some embodiments, the demand publishing time and each supply publishing time are used to calculate the time difference between the demand publishing time and the supply publishing time, and the unit of the time difference can be day, hour, minute or second.

[0044] If the historical evaluation data is numerical information such as good review rate, on-time delivery rate, investment success rate, and past project success rate, the historical evaluation data can be mapped to a time decay coefficient through a first preset mapping relationship. The first preset mapping relationship can be a query table or a piecewise linear function, for example, the first preset mapping relationship can include: when the good review rate is higher than 95%, the decay coefficient is 0.05, and when the good review rate is between 80% and 95%, the decay coefficient is 0.1.

[0045] If the historical evaluation data is product evaluation, investment party evaluation, financing party evaluation, supplier evaluation, etc. text data, a trained sentiment analysis model, such as a model based on BERT architecture, can be used to analyze the text evaluation and determine the positive sentiment numerical score corresponding to the text evaluation, for example, a sentiment score from -1 to 1 or from 0 to 1, and the higher the score, the more positive the sentiment corresponding to the text evaluation. After the quantitative processing of the text data by the trained sentiment analysis model, the average positive sentiment numerical score corresponding to the historical evaluation can be determined, and the average positive sentiment numerical score is mapped to the time decay coefficient through a second preset mapping relationship.

[0046] After determining the time decay coefficient, the time decay factor can be calculated according to the time decay coefficient and the time difference according to the linear decay model, the segmented decay model, the exponential decay model, etc. Time decay factor is negatively correlated with time difference.

[0047] In some embodiments, the linear decay model can be a = 1 - l * t, and the exponential decay model can be a = e^(-l * t), where a is the time decay factor, l is the time decay coefficient, and t is the time difference between the supply publishing time and the demand publishing time. The exponential decay model can smoothly simulate the natural decay process of information value and improve the matching accuracy in time-sensitive scenarios.

[0048] Step 104, calculating the spatial decay factor of the demand information and each supply information according to the demand location information and the supply location information.

[0049] The spatial decay factor is used to quantify the influence of spatial distance on matching degree, and the value decreases with the increase of distance, reflecting the proximity of geographical location.

[0050] In some embodiments, the spatial distance between the demand location and the supply location can be calculated, and then the spatial distance is converted into the spatial decay factor. The spatial decay factor is negatively correlated with the spatial distance.

[0051] Step 105, calculating the target matching degree of the demand information and each supply information according to the time decay factor, the spatial decay factor and the initial matching degree.

[0052] In some embodiments, the time decay factor and the spatial decay factor can be used as weights and combined with the initial matching degree. For example, through weighted sum or product fusion, ensure that time sensitivity and spatial proximity are dynamically included in the matching degree calculation, so as to more accurately calculate the matching degree of the supply information and the demand information.

[0053] In some embodiments, a matching formula Sim=(1-a)*Sim_content+a*(β*Sim_content+(1-β)*Sim_space) can be used to comprehensively consider the influence of content features, time features and space features on the matching degree, where Sim_content is the matching degree based on content features, i.e., the initial matching degree, and Sim_space is the matching degree based on space features. The matching degree based on space features (Sim_space) is used to quantify the degree of matching between the demand location and the supply location in the spatial dimension, and its value range is [0, 1], where 1 represents perfect matching in space, and 0 represents complete mismatch in space.

[0054] In some embodiments, Sim_space=1-(d / D_max).

[0055] When d=0, Sim_space=1, indicating that the locations completely coincide and the spatial matching degree is the highest.

[0056] When d≥D_max, Sim_space=0, indicating that it is beyond the acceptable range and the spatial matching degree is the lowest.

[0057] When 0<d<D_max, Sim_space decreases linearly in the interval (0, 1).

[0058] Where D_max can also be set based on the business type, such as D_max=10km for takeout services and D_max=50km for intracity freight. D_max can be the maximum service distance specified by the user in the demand information. For example, when the user marks "only accept local services", D_max is significantly reduced. D_max can also be determined according to the distribution of successful matching distances of similar businesses.

[0059] Step 106, determining at least one supply information matched by the demand information according to the target matching degree.

[0060] In some embodiments, a preset number of supply information can be determined as the supply information matched by the demand information in the order from high to low according to the target matching degree. The supply information matched by the demand information is displayed to the user so that the user can obtain the desired supply information.

[0061] The supply-demand matching method provided by the present application will be described in detail below by way of examples.

[0062] In the investment scenario, the investor publishes demand information, and the demand description is: investment period 3-5 years, B round artificial intelligence enterprise, publication time January 1, 2025, location information A city, and historical evaluation data including: positive feedback 50 times, negative feedback 10 times.

[0063] The supply description of supply information A is: financing for 3-5 years of product development, B round of artificial intelligence, and the supply release time is December 10, 2024, and the location information is B city. The historical evaluation data includes: 50 positive feedbacks and 10 negative feedbacks.

[0064] The supply description of supply information B is: financing for 3-5 years of market expansion, B round of artificial intelligence, and the release time is January 9, 2025, and the location information is A city. The historical evaluation data includes: 50 positive feedbacks and 10 negative feedbacks.

[0065] The supply description of supply information C is: financing for 3-5 years of technology development, B round of artificial intelligence, and the release time is January 9, 2025. The historical evaluation data includes: 80 positive feedbacks and 5 negative feedbacks.

[0066] According to the demand description and the supply description, the initial matching degrees of the demand description with supply information A, supply information B, and supply information C are calculated. Since all supply descriptions are highly similar to the demand description, the initial matching degrees of the demand description with supply information A, B, and C are all 0.9 calculated using the cosine similarity algorithm.

[0067] According to the historical evaluation data, the supply quality score Q is calculated. Exemplarily, the supply quality score Q = positive feedback number / (positive feedback number + negative feedback number). Therefore, the supply quality score Q of supply information A A = 50 / (50 + 10) ≈ 0.833, the supply quality score Q of supply information B B = 50 / (50 + 10) ≈ 0.833, and the supply quality score Q of supply information C C = 80 / (80 + 5) ≈ 0.941.

[0068] According to the demand release time and the supply release time of supply information A, B, and C, the time differences are calculated as 31 days, 1 day, and 1 day, respectively.

[0069] According to the time difference and the supply quality score, the time decay factor a is calculated. Exemplarily, the time decay factor Therefore, the time decay factors of supply information A, B, and C are 0.0056, 0.846, and 0.943, respectively. Here, supply information A and supply information B have different time decay factors due to different release times, and supply information A has a smaller time decay factor due to a more distant release time. Supply information B and C have different time decay factors due to different historical evaluation data, and supply information C has a larger time decay factor due to supply information B.

[0070] The supply-demand matching method provided by the embodiments of the present application can make the time sensitivity of different supply information different by calculating the time decay factor according to the demand publishing time, the supply publishing time and the historical evaluation data, so that the high-evaluation supply can still maintain a high matching degree as time elapses, can reflect the influence of spatial distance and region on the matching result by calculating the spatial decay factor according to the demand location information and the supply location information, improve the matching accuracy, ensure that the matching result meets the actual demand, reduce the user screening cost and promote the transaction to be reached.

[0071] In some embodiments, in order to further improve the matching accuracy, the step 105 includes steps 201 to 202.

[0072] In step 201, the time decay coefficient of each supply information is calculated according to each historical evaluation data.

[0073] In some embodiments, the historical evaluation data such as user score, click rate or transaction record can be read from the database, and the historical evaluation data is converted into the time decay coefficient λ by a preset algorithm. Moreover, the better the historical evaluation is, the larger the time decay coefficient is, which means that the time decay is slower; the worse the historical evaluation is, the smaller the time decay coefficient is, which means that the time decay is faster.

[0074] In step 202, the time decay factor corresponding to each supply information is calculated according to the time decay coefficient and the time difference between the supply publishing time and the demand publishing time, and the time decay factor is negatively correlated with the time difference.

[0075] The time difference is the time interval between the supply publishing time and the demand publishing time, which is an objective and invariable reference quantity. The time decay coefficient is a dynamic parameter, and the time decay coefficients of different suppliers can be different. The time decay coefficient determines the speed of time decay. In the case that the time decay coefficient is fixed, the larger the time difference is, the smaller the calculated time decay factor is, which can simulate the general rule that the information value decreases with time. In the case that the time difference is fixed, the smaller the time decay coefficient is, the larger the calculated time decay factor is, which can make the supply information of the supplier with high credit, high reliability and good evaluation have higher anti-decay ability.

[0076] The supply-demand matching method provided by the embodiments of the present application can adjust the time decay coefficient through historical evaluation data, and then adjust the time decay factor of different supply information, so as to realize intelligent judgment and differentiated management of the information validity period, so that the supply information with high evaluation can have a longer valid life cycle, and the matching degree of the supply information with low evaluation quickly decays over time. Therefore, while accurately filtering the truly expired information, it is ensured that the high-quality information with actual value will not be misjudged, the matching accuracy in the time-sensitive scene is improved, and the recommendation result is more in line with the real and immediate needs of the user.

[0077] In some embodiments, in order to accurately calculate the time decay coefficient of the supply information, step 201 is refined to include: Step 301, determining the number of positive feedbacks and the number of negative feedbacks in each historical evaluation data; Step 302, calculating the quality score of each supply information based on the number of positive feedbacks and the number of negative feedbacks; Step 303, determining the time decay coefficient corresponding to the supply information according to a plurality of first preset mapping relationships, wherein the time decay coefficient is positively correlated with the quality score.

[0078] The number of positive feedbacks refers to the number of positive evaluations or successful transaction records of the supply information by the user, for example, the number of interested or marked times or the number of completed transactions in the financing investment scene, or the number of full ratings of the supplier in the goods buying and selling scene. The number of positive feedbacks is used to reflect the popularity or reliability of the supply information.

[0079] The number of negative feedbacks refers to the number of negative evaluations or failed transaction records of the supply information by the user, for example, the number of uninterested or failed transactions in the financing investment scene, or the number of complaints of the supplier in the goods buying and selling scene. The number of negative feedbacks is used to reflect the risk or deficiency of the supply information.

[0080] In some embodiments, based on the number of positive feedbacks and the number of negative feedbacks in the historical evaluation data, the quality score Q of each supply information is calculated by using the formula Q=positive feedback number / (positive feedback number+negative feedback number) or the formula Q=(positive feedback number×w1) / (positive feedback number×w1+negative feedback number×w2), and the greater the Q value, the higher the quality of the supply information. Wherein, w1 and w2 are weight factors, which are used to adjust the different weights of the number of positive feedbacks and the number of negative feedbacks in different matching scenes.

[0081] The supply quality score Q is converted into a time decay coefficient λ through a linear function or a nonlinear function, so that high-quality supply information with a large Q value corresponds to a small λ and slow time decay, and low-quality supply information with a small Q value corresponds to a large λ and fast time decay. The linear function can be λ=k×(1-Q), where k is a preset constant. The nonlinear function can be λ=log(1+Q), or Q∈[0.9, 1.0], λ=0.1; Q∈[0.7, 0.9), λ=0.3, and the like.

[0082] The supply-demand matching method provided by the embodiments of the present application can quantitatively evaluate the supply information by calculating the quality score of each supply information based on the positive feedback quantity and the negative feedback quantity, convert the subjective and fuzzy quality evaluation data into an objective and unified numerical index, and then enable the time decay factor of different supply information to be calculated.

[0083] In some embodiments, step 202 is refined to include: Step 401: calculating the time decay factor using the following formula: α=e^(-λ*t), where α is the time decay factor, λ is the time decay coefficient, and t is the time difference between the supply publishing time and the demand publishing time.

[0084] The exponential function is a smooth curve, and its decay process is continuous and has no abrupt change. This ensures that the matching degree does not jump due to a small change in the time difference, thereby improving the stability of the matching result and the user experience. In contrast, linear decay has an abrupt change at the end point, and piecewise decay has a jump at the threshold, both of which can cause severe fluctuations in the matching result. At the same time, α=e^(-λ*t) guarantees that the output α>0 for any non-negative t, which can avoid boundary conditions such as division by zero error or negative values that require additional processing, thereby enhancing robustness.

[0085] In the real world, the value of information usually decays rapidly after publication, and then the decay rate slows down. In the long run, it never goes to zero, but only approaches zero. The curve shape of the exponential decay model α=e^(-λ*t) that is steep at the beginning and gentle at the end can better simulate this natural law, making the time decay factor more consistent with business reality.

[0086] The supply-demand matching method provided by the embodiments of the present application can accurately simulate the natural decay process of information value by using α=e^(-λ*t) to calculate the time decay factor, thereby improving the matching accuracy in scenarios with strong timeliness.

[0087] In some embodiments, the demand location information includes at least one demand location, the supply location information includes at least one supply location, and step 104 is refined to include: Step 501, for each demand location, determine the nearest supply location in the supply information with the minimum spatial distance to the demand location; Step 502, calculate the comprehensive spatial distance of the demand information and each supply information, the comprehensive spatial distance being the sum of the spatial distance of each demand location and the nearest supply location in the supply information; Step 503, calculate the spatial attenuation factor according to the comprehensive spatial distance.

[0088] In some embodiments, for each demand location in the demand information, all supply locations contained in the supply information are traversed, the spatial distance between the demand location and each supply location is calculated using a spatial distance algorithm, such as the Haversine formula based on latitude and longitude, and the minimum value of the spatial distance is selected. The supply location corresponding to the minimum value of the spatial distance is the nearest supply location serving the specific demand location.

[0089] If the demand location information includes one demand location, the spatial distance between the nearest supply location and the demand location is the comprehensive spatial distance of the demand information and the supply information for each supply information.

[0090] If the demand location information includes multiple demand locations, the sum of the spatial distance between the nearest supply location and the demand location is determined as the comprehensive spatial distance of the demand information and the supply information for each supply information.

[0091] In some embodiments, step 503 is refined to include: Step 601, calculate the spatial attenuation factor using the following formula: β = 1 / (1+d), wherein β is the spatial attenuation factor, and d is the comprehensive spatial distance.

[0092] Exemplarily, the demand information A includes three sub-demands, namely, sub-demand 1, sub-demand 2 and sub-demand 3. Sub-demand 1: 100 pieces of equipment, to be delivered to the A warehouse. Sub-demand 2: 80 pieces of equipment, to be delivered to the B store. Sub-demand 3: 120 pieces of equipment, to be delivered to the C warehouse. The demand information A includes three demand locations, namely, A, B and C.

[0093] The supply location information of the supply information A includes one supply location, which is A. The supply location information of the supply information B includes one supply location, which is B. The supply location information of the supply information C includes one supply location, which is C. The supply location information of the supply information D includes two supply locations, which are D and E. The supply location information of the supply information E includes two supply locations, which are F and G.

[0094] For demand location A, in supply information A, the nearest supply location is A, 0 km away; in supply information B, the nearest supply location is B, 1600 km away; in supply information C, the nearest supply location is C, 1200 km away; in supply information D, comparing 700 km of D and 1100 km of E, the nearest supply location is D, 700 km away; in supply information E, comparing 1700 km of F and 1300 km of G, the nearest supply location is G, 1300 km away.

[0095] For demand location B, in supply information A, the nearest supply location is A, 1600 km away; in supply information B, the nearest supply location is B, 0 km away; in supply information C, the nearest supply location is C, 1200 km away; in supply information D, comparing 1100 km of D and 1800 km of E, the nearest supply location is D, 1100 km away; in supply information E, comparing 2100 km of F and 600 km of G, the nearest supply location is G, 600 km away.

[0096] For demand location C, in supply information A, the nearest supply location is A, 1200 km away; in supply information B, the nearest supply location is B, 1200 km away; in supply information C, the nearest supply location is C, 0 km away; in supply information D, comparing 900 km of D and 600 km of E, the nearest supply location is E, 600 km away; in supply information E, comparing 2600 km of F and 1500 km of G, the nearest supply location is G, 1500 km away.

[0097] The comprehensive spatial distance d of each supply information is calculated, according to which, for supply information A, the comprehensive distance d A is 0+1600+1200=2800 km; for supply information B, the comprehensive distance d B is 1600+0+1200=2800 km; for supply information C, the comprehensive distance d C is 1200+1200+0=2400 km; for supply information D, the comprehensive distance d D is 700+1100+600=2400 km; for supply information E, the comprehensive distance d E is 1300+600+1500=3400 km.

[0098] The spatial attenuation factor is calculated using the formula β = 1 / (1+d), and for the supply information A: β ≈ 1 / (1+2800) ≈ 0.000357; for the supply information B: β ≈ 1 / (1+2800) ≈ 0.000357; for the supply information C: β ≈ 1 / (1+2400) ≈ 0.000416; for the supply information D: β ≈ 1 / (1+2400) ≈ 0.000416; and for the supply information E: β ≈ 1 / (1+3400) ≈ 0.000294.

[0099] The supply-demand matching method provided by the embodiments of the present application can intelligently recommend a supplier that can globally optimally serve all warehouses of a purchaser, instead of a local supplier that only serves one warehouse, when the purchaser has multiple warehouses distributed in different cities, thereby solving the problem that the purchaser A needs to sign contracts with, communicate with, pay, and handle logistics of three different suppliers, and reducing the management cost after the intention is reached.

[0100] Based on the supply-demand matching method provided by the above embodiments, the present application further provides a specific implementation of a supply-demand matching device. Please refer to the following embodiments.

[0101] Figure 2 is a structural schematic diagram of a supply-demand matching device provided by another embodiment of the present application. First, referring to Figure 2 The supply-demand matching device 20 provided by the embodiments of the present application includes: The acquisition module 21 is configured to acquire demand information and multiple supply information, wherein the demand information includes demand description, demand publishing time, and demand location information, and the supply information includes supply description, supply publishing time, supply location information, and historical evaluation data. The first calculation module 22 is configured to calculate an initial matching degree of the demand information and each supply information according to the demand description and the supply description. The second calculation module 23 is configured to calculate a time attenuation factor of the demand information and each supply information according to the demand publishing time, each supply publishing time, and each historical evaluation data. The third calculation module 24 is configured to calculate a spatial attenuation factor of the demand information and each supply information according to the demand location information and the supply location information. The fourth calculation module 25 is configured to weight each initial matching degree by using the time attenuation factor and the spatial attenuation factor, to obtain a target matching degree of the demand information and each supply information. The determination module 26 is configured to determine at least one supply information matched to the demand information according to the target matching degree.

[0102] In some possible implementation manners, the second calculation module 23 is specifically configured to: calculate a time attenuation coefficient of each supply information according to each historical evaluation data. According to the time attenuation coefficient and the time difference between the supply publishing time and the demand publishing time, a time attenuation factor corresponding to each supply information is calculated, and the time attenuation factor is negatively correlated with the time difference.

[0103] In some possible implementation manners, the second calculation module 23 can be further configured to: determine the number of positive feedbacks and the number of negative feedbacks in each historical evaluation data; calculate a quality score of each supply information based on the number of positive feedbacks and the number of negative feedbacks; determine a time attenuation coefficient corresponding to the supply information according to a plurality of first preset mapping relationships, and the time attenuation coefficient is positively correlated with the quality score.

[0104] In some possible implementation manners, the second calculation module 23 can be further configured to: calculate the time attenuation factor by using the following formula: α = e^(-λ*t), wherein α is the time attenuation factor, λ is the time attenuation coefficient, and t is the time difference between the supply publishing time and the demand publishing time.

[0105] In some possible implementation manners, the demand position information includes at least one demand position, the supply position information includes at least one supply position, and the fourth calculation module 25 is specifically configured to: for each demand position, determine a nearest supply position with the minimum spatial distance to the demand position from each supply information; calculate a comprehensive spatial distance between the demand information and each supply information, and the comprehensive spatial distance is the sum of the spatial distances between each demand position and the nearest supply position in the supply information; calculate the spatial attenuation factor according to each comprehensive spatial distance.

[0106] In some possible implementation manners, the third calculation module 24 is specifically configured to: for each demand position, determine a nearest supply position with the minimum spatial distance to the demand position from each supply information; calculate a comprehensive spatial distance between the demand information and each supply information, and the comprehensive spatial distance is the sum of the spatial distances between each demand position and the nearest supply position in the supply information; calculate the spatial attenuation factor according to each comprehensive spatial distance.

[0107] In some possible implementation manners, the third calculation module 24 can be further configured to: calculate the spatial attenuation factor by using the following formula: β = 1 / (1+d) Wherein, β is a spatial attenuation factor, and d is a comprehensive spatial distance.

[0108] The various modules of the supply-demand matching device provided by the embodiments of the present application can realize Figure 1 The various steps of the supply-demand matching method provided by the embodiments of the present application can realize the functions of the steps and achieve the corresponding technical effects. For brevity, the functions of the steps will not be described here.

[0109] Figure 3 is a structural schematic diagram of an electronic device provided by another embodiment of the present application. Referring to Figure 3 The supply-demand matching method in the above embodiments, the client exception detection device 30 provided by the embodiments of the present application includes: a processor 31 and a memory 32 storing computer program instructions; and the processor 31 implements any of the supply-demand matching methods in the above embodiments when executing the computer program instructions.

[0110] The supply-demand matching method in the above embodiments, the computer storage medium provided by the embodiments of the present application can be used to implement. The computer storage medium stores computer program instructions; and the computer program instructions are executed by a processor to implement any of the supply-demand matching methods in the above embodiments.

[0111] The embodiments of the present application further provide a computer program product, including a computer program, and the computer program is executed by a processor to implement any of the supply-demand matching methods in the above embodiments.

[0112] It should be further noted that the exemplary embodiments mentioned in the present application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above steps, that is, the steps can be executed in the order mentioned in the embodiments, or different from the order in the embodiments, or several steps are executed simultaneously.

[0113] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other processing device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other processing device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0114] The above is merely specific implementation of the present application, and those skilled in the art can clearly understand the specific working process of the system, module and unit described above for the convenience and brevity of description, which can refer to the corresponding process in the foregoing method embodiments, and will not be described herein. It should be understood that the protection scope of the present application is not limited to this, and any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed in the present application, and these modifications or replacements should be covered within the protection scope of the present application.

Claims

1. A supply-demand matching method characterized by comprising: The method comprises the following steps: obtaining demand information and a plurality of supply information, the demand information comprising a demand description, a demand publishing time and demand location information, the supply information comprising a supply description, a supply publishing time, supply location information and historical evaluation data; calculating an initial matching degree of the demand information and each of the supply information according to the demand description and the supply description; calculating a time decay factor of the demand information and each of the supply information according to the demand publishing time, each of the supply publishing time and each of the historical evaluation data; calculating a spatial decay factor of the demand information and each of the supply information according to the demand location information and the supply location information; calculating a target matching degree of the demand information and each of the supply information according to the time decay factor, the spatial decay factor and the initial matching degree; determining at least one of the supply information matched to the demand information according to the target matching degree.

2. The method of claim 1, wherein, The method of calculating the time decay factor of the demand information and each of the supply information according to the demand publishing time, each of the supply publishing time and each of the historical evaluation data comprises the following steps: calculating a time decay coefficient of each of the supply information according to each of the historical evaluation data; calculating a time decay factor corresponding to each of the supply information according to each of the time decay coefficient and a time difference between the supply publishing time and the demand publishing time, the time decay factor being in a negative correlation with the time difference.

3. The method of claim 2, wherein, The method of calculating the time decay coefficient of each of the supply information according to each of the historical evaluation data comprises the following steps: determining a number of positive feedbacks and a number of negative feedbacks in each of the historical evaluation data; calculating a quality score of each of the supply information based on the number of positive feedbacks and the number of negative feedbacks; determining the time decay coefficient corresponding to each of the supply information according to a plurality of first preset mapping relationships, wherein the time decay coefficient is in a positive correlation with the quality score.

4. The method of claim 2, wherein, The method of calculating the time decay factor corresponding to each of the supply information according to each of the time decay coefficient and the time difference between the supply publishing time and the demand publishing time comprises the following steps: calculating the time decay factor by using the following formula: α = e^(-λ*t), wherein α is the time decay factor, λ is the time decay coefficient, and t is the time difference between the supply publishing time and the demand publishing time.

5. The method of claim 1, wherein, The demand location information comprises at least one demand location, the supply location information comprises at least one supply location, and the method of calculating the spatial decay factor of the demand information and each of the supply information according to the demand location information and the supply location information comprises the following steps: for each of the demand locations, determining a nearest supply location with the minimum spatial distance from the demand location among each of the supply information; calculating a comprehensive spatial distance of the demand information and each of the supply information, the comprehensive spatial distance being a sum of the spatial distances between each of the demand locations and the nearest supply location in the supply information; calculating the spatial decay factor according to each of the comprehensive spatial distance.

6. The method of claim 5, wherein, The method of calculating the spatial decay factor according to each of the comprehensive spatial distance comprises the following steps: The spatial decay factor is calculated using the following equation: β=1 / (1+d)， where β is the spatial decay factor and d is the integrated spatial distance.

7. A supply-demand matching apparatus characterized by comprising: The apparatus comprises: An obtaining module, configured to obtain demand information and a plurality of supply information, the demand information comprising a demand description, a demand publishing time and demand location information, and the supply information comprising a supply description, a supply publishing time, supply location information and historical evaluation data; A first calculating module, configured to calculate an initial matching degree between the demand information and each of the supply information according to the demand description and the supply description; A second calculating module, configured to calculate a time decay factor between the demand information and each of the supply information according to the demand publishing time, each of the supply publishing time and each of the historical evaluation data; A third calculating module, configured to calculate a spatial decay factor between the demand information and each of the supply information according to the demand location information and the supply location information; A fourth calculating module, configured to weight each of the initial matching degrees using the time decay factor and the spatial decay factor to obtain a target matching degree between the demand information and each of the supply information; A determining module, configured to determine at least one of the supply information matched to the demand information according to the target matching degree.

8. An electronic device, comprising: The device comprises a processor and a memory storing computer program instructions; the processor implements the supply-demand matching method of any one of claims 1-6 when executing the computer program instructions.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer program instructions; the computer program instructions are executed by a processor to implement the supply-demand matching method of any one of claims 1-6.

10. A computer program product, characterised in that, The instructions in the computer program product are executed by a processor of an electronic device, so that the electronic device can execute the supply-demand matching method of any one of claims 1-6.

Images