Business opportunity transaction probability evaluation method and computing device

Abstract

The application discloses a business opportunity transaction probability evaluation method and a computing device, comprising: acquiring business opportunity information and a dynamic business opportunity knowledge base of a target business opportunity; performing structured processing and feature extraction on the business opportunity information to generate a business opportunity data set; inputting the business opportunity data set and the dynamic business opportunity knowledge base into a pre-trained large language model for analysis and processing to output transaction key factors related to the target business opportunity; performing multiple evaluations on the transaction key factors through the large language model to generate multiple rounds of evaluation scores, and determining the final scores of each transaction key factor based on a preset score aggregation method; and performing weighted summation on the final scores of the transaction key factors according to a preset classification weight table to output the transaction probability value of the target business opportunity. In this way, the utilization efficiency of business opportunity data and the accuracy of business opportunity transaction probability calculation can be improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent business opportunity prediction technology, and in particular to a method and calculation device for assessing the probability of a business opportunity being closed. Background Technology

[0002] With the development of big data and artificial intelligence technologies, some business opportunity conversion probability calculation models based on data mining and machine learning have emerged. Existing business opportunity transaction probability calculation models still have the following shortcomings: insufficient data utilization, mostly focusing only on data in some dimensions, such as historical transaction data, while ignoring important information such as equipment operation data and market environment data; low accuracy of calculation results, which cannot meet the actual needs of enterprises.

[0003] Therefore, a method and computing device for assessing the probability of a business opportunity being converted are needed to solve the above problems. Summary of the Invention

[0004] The purpose of this application is to provide a method and calculation device for assessing the probability of a business opportunity being closed, which can improve the efficiency of data utilization and the accuracy of calculating the probability of a business opportunity being closed.

[0005] To achieve the above objectives: In a first aspect, embodiments of this application provide a method for evaluating the probability of a business opportunity transaction. The method includes: acquiring business opportunity information and a dynamic business opportunity knowledge base of a target business opportunity; performing structured processing and feature extraction on the business opportunity information to generate a business opportunity dataset; inputting the business opportunity dataset and the dynamic business opportunity knowledge base into a pre-trained large language model for analysis and processing to output key transaction factors related to the target business opportunity; evaluating the key transaction factors multiple times through the large language model to generate multi-round evaluation scores, and determining the final score of each key transaction factor based on a preset scoring aggregation method; and performing a weighted summation of the final scores of the key transaction factors according to a preset classification weight table to output the transaction probability value of the target business opportunity.

[0006] In one embodiment, the business opportunity information includes equipment start-up data, equipment rental data, and business opportunity data; the acquisition of business opportunity information for a target business opportunity includes: acquiring at least one of the equipment information, customer information, and financial information bound to the target business opportunity; and acquiring at least one of the equipment start-up data, equipment rental data, and business opportunity data through an enterprise system and an IoT device platform.

[0007] In one embodiment, the business opportunity information is structured and features are extracted to generate a business opportunity dataset, including: identifying and removing outliers and missing values ​​in the business opportunity information using the isolated forest algorithm to obtain structured multi-dimensional business opportunity data; extracting features from the multi-dimensional business opportunity data; reducing the dimensionality of the multi-dimensional business opportunity data using principal component analysis; and normalizing the dimensionality-reduced data to generate the business opportunity dataset.

[0008] In one embodiment, the business opportunity dataset and the dynamic business opportunity knowledge base are input into a large language model for processing, so as to output the key factors for closing the deal related to the target business opportunity through the large language model. This includes: inputting the business opportunity dataset and the dynamic business opportunity knowledge base into the large language model; performing semantic analysis on the business opportunity dataset through the large language model; and outputting the key factors for closing the deal based on historical factor matching in the dynamic business opportunity knowledge base.

[0009] In one embodiment, the business opportunity dataset and the dynamic business opportunity knowledge base are input into a large language model for processing, so as to output key transaction factors related to the target business opportunity through the large language model. Then, the process includes: comparing the key transaction factors with historical factors in the dynamic business opportunity knowledge base; if the key transaction factors are not present in the dynamic business opportunity knowledge base, adding the key transaction factors to the dynamic business opportunity knowledge base to update the dynamic business opportunity knowledge base.

[0010] In one embodiment, the final scores of the key transaction factors are weighted and summed according to a preset classification weight table to output the transaction probability value of the business opportunity. This includes: obtaining the indicator type of each key transaction factor; querying the classification weight table according to the indicator type to determine the weight parameter of each key transaction factor; calculating the indicator score of each indicator type based on the weight parameter and the final score; summing the indicator scores of each indicator type to generate and output the transaction probability value of the target business opportunity.

[0011] In one embodiment, the indicator types include: market type, device type, and customer type; the step of querying the classification weight table according to the indicator type to determine the weight parameters of each key transaction factor includes: generating a predicted value of the activity of the business opportunity in the future period based on the impact of the scenario and / or time on the target business opportunity using a time series forecasting tool model, and dynamically adjusting the weight parameters of each indicator type in the classification weight table according to the predicted value, wherein when the predicted value is higher than a preset threshold, the weight parameters of the customer type and market type are increased, and the weight parameter of the device type is decreased.

[0012] In one embodiment, the method further includes: comparing the transaction probability value with a preset threshold; when the transaction probability value exceeds the preset threshold, automatically triggering a customer follow-up strategy or generating a priority work order through the enterprise system.

[0013] In one embodiment, the preset score aggregation method includes a mode extraction method; the step of evaluating the key transaction factors multiple times using the large language model to generate multiple rounds of evaluation scores, and determining the final score of each key transaction factor based on the preset score aggregation method, includes: evaluating all key transaction factors related to the target business opportunity multiple times using the large language model; obtaining the multiple rounds of evaluation scores corresponding to each key transaction factor; for each key transaction factor, counting the frequency of each score in the multiple rounds of evaluation scores, and determining the score with the highest frequency of occurrence as the final score of the key transaction factor based on the mode extraction method.

[0014] Secondly, embodiments of this application provide a computing device, including: a processor and a memory storing a computer program, wherein when the processor runs the computer program, the steps of the above-described method are implemented.

[0015] Thirdly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0016] The present application provides a method and computing device for evaluating the probability of a business opportunity transaction. The method includes: acquiring business opportunity information and a dynamic business opportunity knowledge base of the target business opportunity; performing structured processing and feature extraction on the business opportunity information to generate a business opportunity dataset; inputting the business opportunity dataset and the dynamic business opportunity knowledge base into a pre-trained large language model for analysis and processing to output key transaction factors related to the target business opportunity; evaluating the key transaction factors multiple times through the large language model to generate multi-round evaluation scores, and determining the final score of each key transaction factor based on a preset scoring aggregation method; and performing a weighted summation of the final scores of the key transaction factors according to a preset classification weight table to output the transaction probability value of the target business opportunity. This improves the utilization efficiency of business opportunity data and the accuracy of business opportunity transaction probability calculation. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the business opportunity transaction probability assessment method provided in an embodiment of the present invention.

[0018] Figure 2 A schematic diagram illustrating the specific process of the business opportunity transaction probability assessment method provided in this embodiment of the invention.

[0019] Figure 3This is a schematic diagram of the structure of a computing device provided in an embodiment of the present invention. Detailed Implementation

[0020] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. In the following description relating to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements.

[0021] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, components, features, and elements with the same names in different embodiments of this application may have the same meaning or different meanings, the specific meaning of which must be determined by its interpretation in that specific embodiment or further in conjunction with the context of that specific embodiment.

[0022] It should be understood that although the terms first, second, third, etc., may be used herein to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this document, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if," as used herein, can be interpreted as "when," "when," or "in response to determination." Furthermore, as used herein, the singular forms "a," "an," and "the" are intended to also include the plural forms unless the context indicates otherwise. It should be further understood that the terms "comprising," "including," indicate the presence of the stated feature, step, operation, element, component, item, kind, and / or group, but do not exclude the presence, occurrence, or addition of one or more other features, steps, operations, elements, components, items, kinds, and / or groups. The terms "or" and "and / or" as used herein are to be interpreted as inclusive, or mean any one or any combination thereof. Therefore, "A, B, or C" or "A, B, and / or C" means "any one of the following: A; B; C; A and B; A and C; B and C; A, B, and C". Exceptions to this definition will only occur if the combination of elements, functions, steps, or operations is inherently mutually exclusive in some way.

[0023] It should be understood that although the steps in the flowcharts of this application's embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0024] It should be noted that step designations such as S1 and S2 are used in this document for the purpose of more clearly and concisely describing the corresponding content, and do not constitute a substantial limitation on the order. In specific implementation, those skilled in the art may execute S2 first and then S1, etc., but these should all be within the protection scope of this application.

[0025] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0026] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "part," or "unit" may be used interchangeably.

[0027] See Figure 1 This application provides a method that can be executed by a business opportunity transaction probability assessment device provided in this application. This device can be implemented using software and / or hardware. In this embodiment, a computing device is used as the executing entity for this method. The method provided in this embodiment includes: Step S1: Obtain business opportunity information and dynamic business opportunity knowledge base for the target business opportunity.

[0028] Among them, a target business opportunity refers to a potential transaction opportunity in the market environment that arises from supply and demand, technological changes, policy adjustments, or changes in consumption trends, and can satisfy unmet needs through goods or services, ultimately leading to commercial profits.

[0029] The dynamic business opportunity knowledge base is constructed from data such as enterprise business opportunity assessment factors, business opportunity assessment rules, and industry knowledge. Its data is stored in the enterprise's internal system and can be directly accessed when needed.

[0030] The business opportunity information mainly includes three categories: market information, customer information, and financial information. Market information is not stored in the enterprise system and includes equipment operation data and equipment rental data. Additionally, business opportunity information also includes customer and financial information stored in the enterprise system, such as market size, competitive landscape, customer scale, credit rating, and historical transaction amounts.

[0031] It should be noted that business opportunity data refers to business opportunity information other than equipment start-up data and equipment rental data.

[0032] In one embodiment, the business opportunity information includes equipment start-up data, equipment rental data, and business opportunity data; the acquisition of business opportunity information for a target business opportunity includes: acquiring at least one of the equipment information, customer information, and financial information bound to the target business opportunity; and acquiring at least one of the equipment start-up data, equipment rental data, and business opportunity data through an enterprise system and an IoT device platform.

[0033] The process involves obtaining the identification information (such as name and ID) of the target business opportunity, and then using this information to retrieve the target equipment information and target customer information associated with that opportunity. Based on this information, relevant equipment operation data, equipment rental data, and business opportunity data are then obtained from the enterprise system and / or the Internet of Things (IoT). It should be noted that if no equipment information is associated with the target business opportunity, it is not necessary to retrieve equipment operation data and equipment rental data.

[0034] Specifically, customer and financial information is obtained through order information stored in the enterprise system; equipment operation data is mainly obtained through an IoT platform; and equipment rental data is generally not stored internally by the enterprise, but is obtained through sales visits and surveys or third-party leasing platforms. For example, by pre-installing IoT sensors on the equipment, equipment operation data is transmitted back to the enterprise's IoT platform; and equipment rental data is obtained through sales visits and surveys or by acquiring data from industry leasing platforms.

[0035] For example, real-time equipment operation data from the device IoT system, equipment rental data from the leasing platform, and business opportunity data from the enterprise system can be obtained. Among them, equipment operation data may include: equipment start-up frequency, single runtime, regional distribution heat map, etc.; equipment rental data may include: rental payment on-time rate, historical default records, leasing cycle volatility, etc.; business opportunity data may include: market size, competitive landscape, customer scale, credit rating, historical transaction amount, etc.

[0036] Business opportunity data includes: basic customer information, historical order information, external market information, completed business opportunities, and lost business opportunities. Multi-dimensional business opportunity data can be obtained through equipment operation data, equipment rental data, basic customer information, historical order information, and external market information.

[0037] The above method, by collecting multi-dimensional business opportunity data, makes full use of various business opportunity information related to the target business opportunity, such as equipment start-up information and equipment rental information, thereby improving the utilization rate of data.

[0038] Step S2: Perform structured processing and feature extraction on the business opportunity information to generate a business opportunity dataset.

[0039] Among them, the business opportunity dataset consists of multi-dimensional business opportunity data after structured processing and feature extraction of business opportunity information.

[0040] In one embodiment, the business opportunity information is structured and features are extracted to generate a business opportunity dataset, including: identifying and removing outliers and missing values ​​in the business opportunity information using the isolated forest algorithm to obtain structured multi-dimensional business opportunity data; extracting features from the multi-dimensional business opportunity data; reducing the dimensionality of the multi-dimensional business opportunity data using principal component analysis; and normalizing the dimensionality-reduced data to generate the business opportunity dataset.

[0041] Structured processing: Based on the acquired equipment operation and rental data, combined with customer basic information, historical order information, and external market information, outliers and missing values ​​are removed from the above data using the Isolation Forest algorithm, thereby forming structured multi-dimensional business opportunity data. This multi-dimensional business opportunity data includes equipment operation data, equipment rental data, and business opportunity data.

[0042] The Isolation Forest algorithm is used to identify outliers in equipment operation data, rental data, and other data. Examples are as follows: IsolationForest( n_estimators=100, / / Number of random trees generated in the forest max_samples='auto', / / Sampling rules, apply to all samples contamination='auto', / / The proportion of abnormal trees max_features=1.0, / / Set the number of features for each tree to 1 bootstrap=False, / / Samples are not replaced in the sample pool In one implementation, equipment operation data includes: equipment model, material number, operation status, start time, end time, cumulative operation time, daily operation time, and average daily operation time. Equipment rental information includes: product group, equipment model, city (district / county) location, and monthly rental fee. Other business opportunity data includes: market size, competitive landscape, customer scale, credit rating, and historical transaction amount.

[0043] Feature extraction: Principal component analysis (PCA) is used to reduce the dimensionality of the multidimensional business opportunity data and normalize the data to make the data distribution more in line with the normal distribution, resulting in a structured and standardized business opportunity dataset (i.e., the set of all business opportunity data), which is convenient for subsequent large model inference.

[0044] Principal Component Analysis (PCA) is used to reduce the dimensionality of multi-dimensional business opportunity data, forming orthogonal low-dimensional data that facilitates subsequent reasoning and analysis. An example is shown below: PCA(pca_data_sj[, -ncol(pca_data_sj)], ncp =30, graph = FALSE) / / Reduces the number of principal components in the sample data to 30.

[0045] By combining Isolation Forest, PCA, and Normalization, the business opportunity dataset is transformed from a sales report into a reliable decision engine for AI models.

[0046] Step S3: Input the business opportunity dataset and dynamic business opportunity knowledge base into the pre-trained large language model for analysis and processing, so as to output the key factors for closing deals related to the target business opportunity.

[0047] Large Language Models (LLMs) are artificial intelligence models trained on massive amounts of text data using deep learning techniques. They possess powerful natural language understanding and generation capabilities and are widely used in text processing, dialogue systems, and other fields. Examples of large language models include GPT-4o, Gemini, and Qwen / DeepSeek.

[0048] The dynamic business opportunity knowledge base is constructed from enterprise business opportunity assessment factors, business opportunity assessment rules, industry knowledge, etc.

[0049] Before inputting the business opportunity dataset into the large language model, the feature vectors in the dataset need to be mapped to a format compatible with the LLM model input via an adapter. Then, equipment operating data, equipment rental fees, other business opportunity evaluation factors, business opportunity evaluation rules, industry knowledge, etc., are integrated into the LLM model as a knowledge base. The LLM model, combining semantic understanding and contextual analysis, enables causal inference about key factors influencing the success of business opportunities. For example, by calling a large language model fine-tuned for industrial scenarios, semantic parsing and factor deconstruction can be performed on the business opportunity dataset to identify key factors affecting the probability of a successful transaction.

[0050] In one embodiment, the business opportunity dataset and the dynamic business opportunity knowledge base are input into a large language model for processing, so as to output the key factors for closing the deal related to the target business opportunity through the large language model. This includes: inputting the business opportunity dataset and the dynamic business opportunity knowledge base into the large language model; performing semantic analysis on the business opportunity dataset through the large language model; and outputting the key factors for closing the deal based on historical factor matching in the dynamic business opportunity knowledge base.

[0051] For example, key factors for closing a business opportunity are pre-maintained in the knowledge base: factor1, factor2, ..., factorN. These key factors include various indicators and their corresponding definitions. For instance, opportunity effectiveness indicators include quantitative scoring of customer purchasing motivation, customer authenticity, and equipment utilization rate; opportunity risk indicators include quantitative scoring of customer risk level, customer internal and external credit history, and equipment rental fees; and opportunity timeliness indicators include quantitative scoring of differences in entry time with competitors and bidding time. It should be noted that because different data have varying degrees of impact on business opportunities—for example, for the manufacturing industry, a customer's cash flow capability is more important than their creditworthiness—it is necessary to score each key factor for closing the opportunity.

[0052] After inputting the business opportunity data and dynamic business opportunity knowledge base into the large model, the business opportunity information to be analyzed for this business opportunity (i.e., the target business opportunity) is processed to obtain the key factors for closing the deal: newFactor1, newFactor2, ..., newFactorN. Example prompt input to the large language model: Referring to the key factors for closing deals (factor1, factor2, ..., factorN) in the existing knowledge base, analyze whether the business opportunity dataset obtained for this business opportunity contains all the key factors for closing deals in the knowledge base (output the included key factors for closing deals; if there are any new key factors for closing deals outside the knowledge base, output them as well). Output the key factors for closing deals {newFactorN} that may be involved in this business opportunity (including information on closed deals and lost deals). It allows adding key factors for closing deals based on the dynamic business opportunity knowledge base, but the added key factors for closing deals need to be saved to the dynamic business opportunity knowledge base to achieve dynamic updates of the knowledge base.

[0053] By using large LLM models for feature extraction and causal reasoning, we can uncover deep-seated features and causal relationships in the data, making the model more applicable.

[0054] In one embodiment, the business opportunity dataset and the dynamic business opportunity knowledge base are input into a large language model for processing, so as to output key transaction factors related to the target business opportunity through the large language model. Then, the process includes: comparing the key transaction factors with historical factors in the dynamic business opportunity knowledge base; if the key transaction factors are not present in the dynamic business opportunity knowledge base, adding the key transaction factors to the dynamic business opportunity knowledge base to update the dynamic business opportunity knowledge base.

[0055] For example, check whether there is a non-inclusion relationship between the set of key transaction factors {newFactorN} output by the large language model and the set of {factorN} in the knowledge base; if so, update the set of {factorN} to supplement and enrich the set of transaction factors and perform knowledge iteration; the set of {factorN} is the largest set, and the set of {newFactorN} for different business opportunities may differ.

[0056] Step S4: The key factors for transaction are evaluated multiple times using a large language model to generate multiple rounds of evaluation scores, and the final score of each key factor for transaction is determined based on a preset score aggregation method.

[0057] The preset scoring aggregation methods include mode extraction, average calculation, and median selection. This implementation prioritizes mode extraction, which selects the score with the most occurrences of the same score from multiple batches of evaluation scores as the final score for that key transaction factor. If there is no mode among the multiple rounds of evaluation scores for a key transaction factor, methods such as average calculation and median selection can be used to calculate the average or select the median as the final score for each key transaction factor.

[0058] In one embodiment, the key factors for transaction are evaluated multiple times using a large language model to generate multiple evaluation scores, and the final score of each key factor for transaction is determined based on a preset score aggregation method. This includes: evaluating all key factors for transaction related to the target business opportunity multiple times using a large language model; obtaining the multiple evaluation scores corresponding to each key factor for transaction; for each key factor for transaction, counting the frequency of each score in the multiple evaluation scores, and determining the score with the highest frequency of occurrence as the final score of the key factor for transaction based on the mode extraction method.

[0059] Optionally, determining the final score of each of the key transaction factors based on a preset scoring aggregation method further includes at least one of the following: calculating the average of the multiple rounds of evaluation scores for each key transaction factor and using the calculation result as the final score; taking the median value of the multiple rounds of evaluation scores for each key transaction factor and using the median value as the final score.

[0060] Specifically, the large language model evaluates and scores all transaction key factors / newFactors for the same business opportunity multiple times (no less than 3 times), defining a score (newFactor key factor, grade score), thereby obtaining multiple batches of evaluation scores V1{score1, score2, ..., scoreN}, V2{score1, score2, ..., scoreN}, ..., Vm{score1, score2, ..., scoreN}.

[0061] For each key transaction factor / newfactorX, calculate the evaluation score scoreX for different batches. Then, select the score with the most frequent occurrences of the same score across all batches as the final score / score for this key transaction factor / newfactorX. Repeat this process for all key transaction factors newfactor, and finally combine them to obtain the final score {score1} for each key transaction factor in this opportunity / target opportunity. score2 ... scoreN }

[0062] For example, after evaluation, the large language model outputs four batches of evaluation scores for two key transaction factors: V1{0.2,0.3}, V2{0.1,0.4}, V3{0.2,0.5}, and V4{0.3,0.4}. For each key transaction factor, the score with the most frequent occurrences among the scores for the same key transaction factor in different batches can be used as the final score for that factor. In the example, the most frequent score in the first key transaction factor / newFactor1 is 0.2, and the most frequent score in the second key transaction factor / newFactor2 is 0.4. Finally, the final scores of these two key transaction factors are concatenated to form the final score of the key transaction factor. In the example, the final score of the key transaction factor for the target business opportunity should be {0.2,0.4}. It should be noted that the average calculation method and the median value method can be referred to the above example, so they will not be described in detail here.

[0063] By using the above method, a large language model is used to score all key transaction factors in multiple rounds, and the factor with the highest frequency is taken as the final score, which can eliminate model illusion as much as possible.

[0064] Step S5: Calculate the weighted sum of the final scores of the key factors for the transaction based on the preset classification weight table to output the transaction probability value of the target business opportunity.

[0065] The preset classification weight table refers to dividing multiple key transaction factors into major categories; for example, dividing key transaction factors into categories such as customer information, equipment information, and market information; and then setting corresponding weight parameters for each category. The weight parameters corresponding to different business opportunities may be different for each category.

[0066] In one embodiment, the final scores of the key transaction factors are weighted and summed according to a preset classification weight table to output the transaction probability value of the business opportunity. This includes: obtaining the indicator type of each key transaction factor; querying the classification weight table according to the indicator type to determine the weight parameter of each key transaction factor; calculating the indicator score of each indicator type based on the weight parameter and the final score; summing the indicator scores of each indicator type to generate and output the transaction probability value of the target business opportunity.

[0067] For example, market category indicator A {score1, score2, score3} is assigned a weight of 1, customer category indicator B {score4, score5, score6} is assigned a weight of 2, and equipment category indicator C {score7, score8, score9} is assigned a weight of 3, thereby calculating the score for each category.

[0068] Among them, the probability of a business opportunity being closed = weight 1 A+ weight 2 B+ weight 3 C; A = score1 + score2 + score3.

[0069] In this implementation, the transaction probability value is a range, such as [0-20%], [21%-40%], [41%-60%], [61%-80%], [81%-100%] (this range can be customized according to actual conditions, such as [61%-70%], [60%-90%], etc.). For example, when the final scores of each category are weighted and summed to calculate a transaction probability of 68%, the transaction probability value is output as [61%-80%].

[0070] The indicator types include: market type, equipment type, and customer type.

[0071] In one embodiment, the step of querying the classification weight table according to the indicator type to determine the weight parameters of each key transaction factor includes: generating a predicted value of the activity of the business opportunity in the future period based on the impact of the scenario and / or time on the target business opportunity using a time series forecasting tool model, and dynamically adjusting the weight parameters of each indicator type in the classification weight table according to the predicted value. Specifically, when the predicted value is higher than a preset threshold, the weight parameters of the customer type and market type are increased, and the weight parameter of the device type is decreased.

[0072] The preset classification weight table can be dynamically updated according to industry categories, and the weight allocation is adaptively adjusted based on the regression analysis results of historical transaction data.

[0073] The Prophet time series forecasting tool is used to adjust weighting indicators based on the impact of time and important holidays on business opportunities. An example is shown below: model = Prophet(holidays=holidays_df) model.fit(df) future = model.make_future_Sjdata(periods=30) forecast = model.predict(future) By using the LLM model to adjust feature weights according to different customer scenarios and market dynamics, the probability of closing a business opportunity is calculated, thus improving the accuracy of the business opportunity closing probability calculation.

[0074] In one embodiment, the business opportunity conversion probability assessment method further includes: comparing the conversion probability value with a preset threshold; when the conversion probability value exceeds the preset threshold, automatically triggering a customer follow-up strategy or generating a priority work order through the enterprise system.

[0075] When the probability of a sale exceeds a preset threshold (e.g., 60%), the enterprise system will automatically trigger a customer follow-up strategy or generate a priority work order to remind business personnel to follow up on the business opportunity.

[0076] Based on the same inventive concept as the foregoing embodiments, the foregoing embodiments will be described in detail below through a specific example. For example... Figure 2 As shown, after inputting the identifier of the target business opportunity, the system automatically retrieves relevant opportunity information and a dynamic business opportunity knowledge base. The opportunity information is categorized using a large language model to obtain equipment information, customer information, and market information. These three information are then analyzed separately to identify key factors influencing the success of the target business opportunity. The large language model then quantifies and scores each key factor, resulting in equipment scores, customer scores, and market scores. Finally, the equipment scores, customer scores, and market scores are summed, and the range within which the sum falls is output as the probability value of the successful transaction for the target business opportunity.

[0077] In summary, the method provided in the above embodiments collects multi-dimensional business opportunity information, makes full use of various business opportunity-related information, and then uses an LLM large model to extract features and infer causal relationships, thereby mining the potential deep features and causal relationships in the data.

[0078] Based on the same inventive concept as the foregoing embodiments, this embodiment of the invention provides a computing device, such as... Figure 3 As shown, the device includes: a processor 310 and a memory 311 storing a computer program; wherein, Figure 3 The processor 310 shown in the diagram does not indicate that there is only one processor 310, but only indicates the positional relationship of the processor 310 relative to other devices. In practical applications, there can be one or more processors 310; similarly, Figure 3 The memory 311 illustrated herein has the same meaning, that is, it is only used to indicate the positional relationship of memory 311 relative to other devices. In practical applications, there can be one or more memories 311. When the processor 310 runs the computer program, the method applied to the above-mentioned device is implemented.

[0079] The device may also include at least one network interface 312. The various components of the device are coupled together via a bus system 313. It is understood that the bus system 313 is used to implement communication between these components. In addition to a data bus, the bus system 313 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 3 The general designated all buses as Bus System 313.

[0080] The memory 311 can be volatile memory or non-volatile memory, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memory 311 described in the embodiments of the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0081] The memory 311 in this embodiment of the invention is used to store various types of data to support the operation of the device. Examples of this data include: any computer programs used to operate on the device, such as operating systems and applications; contact data; phonebook data; messages; pictures; videos, etc. The operating system includes various system programs, such as the framework layer, core library layer, driver layer, etc., used to implement various basic services and handle hardware-based tasks. Applications can include various applications, such as media players, browsers, etc., used to implement various application services. Here, the program implementing the method of this embodiment of the invention can be included in the application.

[0082] Based on the same inventive concept as the foregoing embodiments, this embodiment also provides a computer-readable storage medium storing a computer program. The computer-readable storage medium can be a magnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM), etc.; it can also be various devices including one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc. When the computer program stored in the computer-readable storage medium is run by a processor, it implements the above method. For the specific steps implemented when the computer program is executed by the processor, please refer to [link to relevant documentation]. Figure 1 The description of the illustrated embodiments will not be repeated here.

[0083] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0084] In this document, the terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, which includes not only the elements listed but also other elements not expressly listed.

[0085] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for assessing the probability of a business opportunity being closed, characterized in that, include: Obtain business opportunity information and a dynamic business opportunity knowledge base for target businesses; The business opportunity information is structured and its features are extracted to generate a business opportunity dataset; The business opportunity dataset and dynamic business opportunity knowledge base are input into a pre-trained large language model for analysis and processing to output key factors for closing deals related to the target business opportunity. The key transaction factors are evaluated multiple times using the large language model to generate multiple rounds of evaluation scores, and the final score of each key transaction factor is determined based on a preset score aggregation method. The final scores of the key factors for the transaction are weighted and summed according to a preset classification weight table to output the transaction probability value of the target business opportunity.

2. The business opportunity transaction probability assessment method as described in claim 1, characterized in that, The business opportunity information includes equipment start-up data, equipment rental data, and business opportunity data; The business opportunity information for obtaining the target business opportunity includes: Obtain at least one of the equipment information, customer information, and financial information associated with the target business opportunity; Obtain at least one of the following: equipment operation data, equipment rental data, and business opportunity data through enterprise systems and IoT device platforms.

3. The business opportunity transaction probability assessment method as described in claim 1 or 2, characterized in that, The process of structuring and extracting features from the business opportunity information to generate a business opportunity dataset includes: The isolated forest algorithm is used to identify and remove outliers and missing values ​​from the business opportunity information to obtain structured, multi-dimensional business opportunity data. Feature extraction is performed on the multi-dimensional business opportunity data, and then principal component analysis is used to reduce the dimensionality of the multi-dimensional business opportunity data. The dimensionality-reduced data is then normalized to generate the business opportunity dataset.

4. The business opportunity transaction probability assessment method as described in claim 1, characterized in that, The step of inputting the business opportunity dataset and dynamic business opportunity knowledge base into a large language model for processing, so as to output key transaction factors related to the target business opportunity through the large language model, includes: Input the business opportunity dataset and the dynamic business opportunity knowledge base into the large language model; The business opportunity dataset is semantically analyzed using the large language model, and based on historical factor matching in the dynamic business opportunity knowledge base, the key factors that influence the probability of closing a business opportunity are output.

5. The business opportunity transaction probability assessment method as described in claim 4, characterized in that, The process involves inputting the business opportunity dataset and dynamic business opportunity knowledge base into a large language model for processing, so as to output key transaction factors related to the target business opportunity through the large language model, followed by: The key factors for the transaction are compared with the historical factors in the dynamic business opportunity knowledge base; If the key transaction factor is not present in the dynamic business opportunity knowledge base, the key transaction factor is added to the dynamic business opportunity knowledge base to update the dynamic business opportunity knowledge base.

6. The business opportunity transaction probability assessment method as described in claim 1, characterized in that, The step of weighting and summing the final scores of the key transaction factors according to a preset classification weight table to output the transaction probability value of the business opportunity includes: Obtain the indicator types for each key transaction factor; Query the classification weight table according to the indicator type to determine the weight parameters of each key transaction factor; Based on the weight parameters and the final score, calculate the index score for each index type; The scores of each indicator type are summed to generate and output the probability value of closing a transaction for the target business opportunity.

7. The business opportunity transaction probability assessment method as described in claim 6, characterized in that, The indicator types include: market type, equipment type, and customer type; The step of querying the classification weight table according to the indicator type to determine the weight parameters of each key transaction factor includes: Using a time series forecasting tool model, based on the impact of scenario and / or time on target business opportunities, a predicted value of business opportunity activity in future periods is generated. The weight parameters of each indicator type in the classification weight table are dynamically adjusted according to the predicted value. When the predicted value is higher than a preset threshold, the weight parameters of the customer type and market type are increased, while the weight parameter of the device type is decreased.

8. The business opportunity transaction probability assessment method as described in claim 1, characterized in that, The preset score aggregation method includes the mode extraction method; The process of evaluating the key transaction factors multiple times using the large language model to generate multi-round evaluation scores, and determining the final score for each key transaction factor based on a preset score aggregation method, includes: The large language model is used to evaluate all key factors related to the target business opportunity multiple times. Obtain the multi-round evaluation scores for each key transaction factor; For each of the aforementioned key factors for transaction, the frequency of each score in the multiple rounds of evaluation is counted, and the score with the highest frequency is determined as the final score of the key factor for transaction based on the mode extraction method.

9. The business opportunity transaction probability assessment method as described in claim 1, characterized in that, The method further includes: The transaction probability value is compared with a preset threshold. When the probability of a transaction exceeds the preset threshold, the enterprise system automatically triggers a customer follow-up strategy or generates a priority work order.

10. A computing device, characterized in that, include: The processor and the memory storing a computer program implement the steps of the business opportunity transaction probability assessment method according to any one of claims 1 to 8 when the processor runs the computer program.

Images