TW201503033A - A system and method using multi-dimensional rating to determine an entity's future commercial viability - Google Patents

Abstract

A method and system for determining an entity's future commercial viability which comprises: (a)using a first predictive modeling, determining a future commercial viability of the entity, the first predictive modeling is derived by identifying patterns in data and relating to predictive attributes, thereby generating a viability score; (b) using predictive modeling to generate a relative ranking of the entity against its peer group, thereby generating a comparative viability score (i.e., portfolio comparison); (c) measuring data depth to quantify how much is known about the entity and, thus, how much confidence we have in the viability score and comparative viability score, thereby generating a data depth indicator; (d) assigning a company profile by segmentation to define and group the entity with other similar entities in terms of size, years in business, availability of complete financial statement and commercial trade history; and (e) outputting a multi-dimensional viability rating comprising the viability score, comparative viability score, data depth indicator, and company profile.

Description

System and method for determining the future commercial viability of an entity using a multi-dimensional rating system Cross-reference application

The present application claims priority to U.S. Provisional Application No. 61/818,729, filed on May 2, 2013, which is hereby incorporated by reference.

Field of invention

This disclosure is generally concerned with predictive and descriptive scoring/analysis. The Survival Rating according to this disclosure expresses a multi-dimensional rating of an insightful and highly reliable assessment of the entity's future business activity. The predictive component predicts the likelihood that a company will fail, become inactive, or file for bankruptcy within a certain period of time (eg, the next twelve (12) months). The descriptive component provides an indication of the amount of predictive data that can be used to make a reliable risk assessment, as well as insight into the characteristics of the business (eg, age, type, and size of the business).

Background of the invention

The uniqueness of the viability rating according to this disclosure is that it utilizes the activity of being unable to confirm or dormant (in this document) One step is called "UTC" as part of the dependent/target variable for model development. This is one of the usage scenarios we have defined for the information generated by the assessment of the activity of the business. An enterprise designated as UTC has been dormant for a certain period of time, such as 12 months, and has been found to be inactive by applying various business rules. These rules include, but are not limited to, invalid corporate addresses, unreachable calls, or no trade activity. We previously made this determination using bankruptcy or known bankruptcy. By using UTC attributes, the methods and systems of the present disclosure are capable of identifying a greater number of businesses that are inactive or dormant (and therefore have no signal to confirm their presence). In accordance with the present disclosure, the use of UTC attributes provides an earlier signal related to inactivity or hibernation of the enterprise, rather than relying on severe fault information.

The present disclosure also provides a number of additional advantages that will become apparent as described below.

Summary of invention

The multi-dimensional viability rating includes multiple components; in this example of the disclosure, the viability rating is described as using four (4) components. The first two components largely predict whether an entity will cease to exist, become dormant, or become inactive over the next twelve months. The third component reveals the depth of available data, and the fourth component provides a description of the company from a demographic perspective.

A method and system for determining future commercial viability of an entity, comprising: (a) using predictive modeling to determine future viability of the entity, the predictive modeling being by identifying data (eg, UTC data) type And deriving in association with the predicted attributes to produce a viability score; (b) using predictive modeling to generate a relative ranking of the entity relative to one of its peer groups, thereby generating a comparative viability score; (c) Measuring the depth of the data to quantify the degree of understanding of the entity, and thus quantifying the degree of trust in the viability score and the comparative viability score, thereby generating a data depth indicator; (d) assigning a company profile by Segments define the entity based on a number of characteristics and group the entity with other similar entities, and such features are defined, for example, in terms of size, years of operation, availability of complete financial statements, and commercial trade history; and (e A multi-dimensional viability rating containing the viability score, comparative viability score, data depth indicator, and company profile is output.

The viability score is a predictive rating on a viability score scale, where the range is, for example, between about 1 and about 9, and the 1 series of entities collapses or becomes inactive over a period of time compared to other firms. The lowest probability, and the highest probability that the 9 series will close or become inactive.

An exemplary comparative viability score is a predictive rating on a comparative survivability score scale, wherein the range is, for example, between about 1 and about 9, and the 1 series is compared to other firms within the same model segment at a time The lowest probability of collapse or becoming inactive within a segment, and the highest probability that the 9 Series will fail or become inactive.

An exemplary data depth indicator is based on a descriptive rating of a data depth indicator scale, such as between about A and M. Assign A to G on the "class report card" scale, where A is assigned to the enterprise with the highest level of forecast data, the forecast data is selected from the following groups Groups: complete corporate identity data (such as number of employees or industry), extensive commercial trade activities, integrated financial attributes, and mixtures thereof; and assign G to companies with the lowest level of forecasting information. The forecast data is basic identity data. H to M is a special category that prioritizes A to G ratings, giving the user further insight when it is confirmed that the establishment has encountered one of a set of predefined risk conditions.

An exemplary company profile is based on a descriptive rating of a company profile scale, where the range is, for example, between about A and Z. A can represent the largest, longest-established company, and the smallest and youngest company in the X department.

A computer readable storage medium containing executable computer program instructions that, when executed, cause a processing system to perform a method for determining future commercial viability of an entity, the method comprising: (a) using predictions Modeling to determine the future viability of the entity, the prediction model is derived by identifying the type of the data and correlating with the predicted attributes to generate a viability score; (b) using predictive modeling to generate the entity relative Relative ordering of one of its peer groups, resulting in a comparative viability score; (c) measuring the depth of the data to quantify the level of knowledge of the entity, and thus quantifying the trust score and the comparative viability score Degree, which in turn generates a data depth indicator; (d) assigns a company profile that defines the entity based on a number of features by segmentation and groups the entity with other similar entities, and such features are Defined in terms of size, length of operation, availability of complete financial statements, and history of commercial trade; and (e) output one containing the viability score, ratio Viability scores, and company profile data in depth as much as dimension indicator viability rating.

A computer system for determining future commercial viability of an entity, the system comprising a processor executing the following steps stored in the memory; the steps comprising: (a) using predictive modeling to determine the The future viability of the entity, which is derived by identifying the type of the data and correlating with the predicted attributes to produce a viability score; (b) using predictive modeling to generate the entity relative to its peer group One of the relative rankings, which in turn produces a comparative viability score; (c) measures the depth of the data to quantify the degree of knowledge of the entity, and thus quantifies the degree of trust in the viability score and the comparative viability score, thereby generating a a data depth indicator; (d) assigning a company profile that defines the entity based on a number of features by segmentation and grouping the entity with other similar entities, such as size, length of business, , the availability of complete financial statements and the history of commercial trade; and (e) output one containing the viability score, comparative viability score, data Company profile and indicators of many dimensions viability rating.

A computer system for determining future commercial viability of an entity, the system comprising: a database containing activity signal data; and an activity signal generator that uses a plurality of data sources to aggregate the activity signal data The source of the data is from a plurality of companies that do business with an entity of interest; and a model generator that generates a viability score based on a statistical model, wherein a dependent variable performance system uses statistical probability from independent variables Export, these independent variables are created from multiple sources.

The processor executes the following steps stored in the memory; the steps include: (a) using a first predictive modeling to determine that one of the entities is not To commercial viability, the first predictive modelling is derived by identifying the type of data and correlating with the predictive attributes to generate a viability score; (b) using a second predictive modelling to generate the entity relative to One of the peer groups is relatively ordered, thereby generating a comparative survivability score; (c) measuring the data depth to quantify the degree of understanding of the entity and the degree of trust in the viability score and the comparative viability score, thereby generating a data depth indicator; (d) assigning a company profile that defines the entity by segmentation and grouping the entity with other similar entities; and (e) outputting one containing the viability score, the comparison The viability score, the data depth indicator, and the multi-dimensional viability rating of the company profile.

The activity signal generator includes: a matching process that generates a signal when a match is found; a login process that receives the signal and inputs the signal into the metadata; and an aggregator from which the metadata The data is aggregated to generate the active signal data. The signal includes at least one signal selected from the group consisting of: (a) identification of the source, data received from the source; (b) time to form the match; (c) Unique identifier 341; and (d) credibility code.

Further objects, features, and advantages of the present disclosure will be understood by reference to the following drawings and detailed description.

100‧‧‧ system

105‧‧‧ computer

110‧‧‧User interface

115‧‧‧ processor

120‧‧‧ memory

125‧‧‧Processing module

129‧‧‧UTC

130‧‧‧ Account receivables

135‧‧‧Detailed trade information

140‧‧‧Corporate References

145‧‧‧Source

145-1~145-N‧‧‧Source

150‧‧‧Network

160‧‧‧Active signal data

165‧‧‧ score

199‧‧‧Storage device

200‧‧‧ method

202, 212, 214, 216, 218, 220, 504~514‧ ‧ steps

205‧‧‧ASD generator

210‧‧‧A/R processing/steps

215‧‧‧Model Generator

220‧‧‧Score process/step

301‧‧‧ descriptor

305‧‧‧ Matching process

306‧‧‧ signal

310‧‧‧ Login process

312‧‧ cycle

313‧‧‧

315‧‧‧Aggregator

320‧‧‧ yuan data

330‧‧‧ unique identifier

335‧‧‧Number of signals

336‧‧‧Reliability code matching

340‧‧ record

502‧‧‧Module

516‧‧‧ viability score

522‧‧‧data depth score

1A is a block diagram of a system for use in the techniques disclosed herein; FIG. 1B is a block diagram of a processing module of the system of FIG. 1A; FIG. 1C is a block diagram of an activity signal generator, the generator is FIG. The components of the processing module; FIG. 2 is a flow chart depicting the scoring process according to the present disclosure, which is used to determine both the viability score and the comparative viability score in the predictive model; The data depth table used in the disclosure; Figure 4 is a company profile used to interpret the portfolio component of the disclosure; Figure 5 is a diagram showing how the survivability score and the data depth score are used to calculate across the four model segments ( That is, a flow chart of the viability rating of the financial section, established trade payments, limited trade payments, and no trade payments; and Figure 6 is an example of a weighting scheme in accordance with the present disclosure.

Detailed description of the preferred embodiment

The Survival Rating is a multi-dimensional rating that expresses an insightful and highly reliable assessment of the company's future activity. Survivability ratings both unpredictable components and descriptive components. Predictive component predictions The likelihood that a company will fail, become inactive, or file for bankruptcy, for example, for a defined period of time (eg, the next 12 months). The descriptive component provides an indication of the amount of predictive data that can be used to make a reliable risk and/or business activity assessment, as well as an insight into firm size measurements based on a range of characteristics, such as the age, type, and size of the business. The exemplary components used to generate the viability rating are:

• Survivability score: a predictive rating on a scale, for example, the scale is in the range of between about 1 and 9, where 1 is compared to other companies over a period of time (eg, the next 12) The lowest probability of failure or becoming inactive within the month, and the highest probability that the 9 Series will fail or become inactive. The UTC 129 data is used as part of the reliable variables in the development of statistical models. UTC 129 data for information on inactive and dormant businesses. The detailed trade data 135 predictor is also an independent variable that is important in model development.

• Portfolio comparison: A predictive rating on a scale, for example, the scale is in the range of between 1 and 9, with 1 firm compared to other firms for a period of time (eg, the next 12) The lowest probability of failure or becoming inactive within the month, and the highest probability that the 9 Series will fail or become inactive. Detailed trade information 135 is used to define model segments that enable the ability to express models that compare the viability of firms within the same level of business activity (eg, firms with low payment transaction counts).

• Data depth indicator: A descriptive rating on a scale, for example, the scale is in the range between approximately A and M. Assign A to G on the “Class Report Card” scale, where for example, assign A to the company with the highest level of forecast data, the highest level of forecast data includes complete corporate identity data, extensive commercial trade activities, and comprehensive financial attributes; The G is assigned to the company with the lowest level of forecast data, and the lowest level of forecast data only includes the basic identity data. Categories such as H to M can be dedicated to special categories that are prioritized over A to G ratings, giving the user further insight when it is confirmed that the enterprise has encountered one of a set of predefined risk conditions. A number of sources are used to define the data depth indicator. Some of the attributes derived from UTC 129, detailed trade data 135, and corporate reference 140 are important participants in the creation of the viability rating component data depth indicator.

• Company Profile: A descriptive rating on a scale, for example, the scale is between about A and Z, with A being the largest, longest-established company, and Z being the smallest, youngest company. The exemplary company profile is defined using a plurality of sources of information including detailed trade information 135 (eg, number of payment transactions) and corporate reference material 140 (eg, years of operation).

The viability rating uses statistical probabilities to classify an enterprise as, for example, a 1 to 9 risk rating segment. Such classifications are based, for example, on the likelihood that a company will fail, become inactive, or file for bankruptcy over a period of time (eg, the next 12 months).

The data depth indicator uses a point system to assign values to data attributes based on the ability of the material attributes to improve the prediction accuracy of the viability rating. The higher the predictability of the data attributes, the more points are assigned. For example, financial information and extensive trade data may have higher predictive indices, resulting in robust forecasts. Therefore, the data receives a higher number of points, thereby placing the company higher on the A to M scale.

Company profiles use segmentation to define businesses that are similar in terms of size (eg, employee and annual sales) and age (eg, business years) and group them into groups.

Survivability ratings leverage the combined capabilities of a wide range of information, including but not limited to corporate activity signals, detailed business transaction payment experiences derived from invoice level data from account receivables.

Survivability ratings use statistical model construction techniques, and these technology packages This includes, but is not limited to, segmentation analysis and subsequent regression analysis.

Exemplary viability scores and portfolio comparisons use statistical probabilities to classify firms as, for example, risk ratings ranging between 1 and 9, with 1 showing the lowest probability of becoming inactive and the 9 being becoming inactive. Probability. These classifications are based on the likelihood that the company will close down, become inactive or file for bankruptcy in the next 12 months.

The statistical probability is established using a statistical model development method (regression), in which the probability of a result (eg, closing or becoming dormant in the next 12 months) is via a model that captures the independent variables (predictors) of this behavior. Observe and observe.

The data depth indicator uses a point system to assign values to data attributes based on the ability of the material attributes to improve the predictive accuracy of viability scores and portfolio comparisons. The higher the predictability of the data attributes, the more points are assigned. For example, financial information and extensive transaction payment information may have a higher forecasting index, resulting in robust forecasts. Therefore, the data receives a higher number of points, thereby placing the company higher on the A to M scale.

The exemplary company profile uses segmentation to define companies that are similar in size (employee and annual sales), age (years of operation), and the availability of complete financial statements, as well as commercial trade history, and group them into groups.

Survivability ratings use multiple sources of information, such as: Corporate Activity Signaling Data (ASD) 160; detailed commercial payment experiences derived from account receivables transaction payment data, which captures monthly trends and is referred to in this document as detailed Trade 135; UTC 129; and Corporate Reference 140. Survival review Levels, for example, predict the likelihood that an enterprise will:

● Voluntary or involuntary closure

● Become dormant or inactive

● Apply for bankruptcy

The underlying model for viability ratings is based on the observational characteristics of thousands of firms and the relationship of these characteristics to the probability of meeting the definitions above.

The model assigns a score, for example, in the range between about 1 and 9. This section divides the scoring range into nine distinct risk groups, one (1) indicating the company with the lowest probability of failure, becoming inactive or filing for bankruptcy, and nine (9) indicating the most probable enterprise. . For example, using this extended definition of an active business, we can predict a business failure for small businesses that may slowly reduce activity over time until they no longer exist.

The data depth indicator provides insight into the level of available forecasting material elements for the enterprise. It allows the user to understand and trust the underlying data input used to assess viability. The key to an exemplary data depth indicator is understood with reference to FIG.

Based on a combination of the following characteristics, the exemplary profile of the company profile category is in the range of A to Z, such as the number of years of operation, the number of employees or the annual sales volume and the amount of payment transactions, ie:

●Young: The business period is less than 5 years

● Established for a long time: the business life is more than 5 years

● Small: less than 10 employees or missing actual employees, or annual sales less than $100,000 or lack of actual sales

● Medium: Between 10 and 49 employees, or annual sales between $100,001 and $499,999

● Large: Employees greater than 50, or annual sales greater than $500,000

● Financial statements are available or unavailable

● Three (3) or more trade payment references available

The company with A profile is the largest and longest-established company with complete financial statements and trade payment information. Company B with the X Profile is the smallest and youngest company with financial information or trade payment information not available. Refer to Appendix B of Figure 4 for an exemplary company profile category.

Model development

The predictive component of the viability rating is based on statistical modeling techniques to select and weight data elements that best predict the relatives of corporate failures, inactivity and bankruptcy, and corporate behavior. The resulting model is a mathematical equation consisting of a series of variables and coefficients (weights) that are calculated for each variable. One of the techniques on which the predictive model is based is the logistic regression technique, which establishes the best practice of establishing a model with binary dependent variables.

Extensive data analysis is performed to determine statistically the variables that are the most important factors for predicting collapse, inactivity, and bankruptcy, and the appropriate weights are calculated for each variable. Hundreds of predictive variables were identified by assessing the combination of “good” companies in the database and those with “bad” performance.

This disclosure utilizes active signal data (ADS) generated by a rule-driven data collection and maintenance system of data sources. ADS is especially beneficial To distinguish between low-risk and high-risk small businesses, small businesses often have limited commercial trade history or no commercial trade history. We have also increased the depth of the information utilized by the scores by using detailed transaction payment information for companies with established commercial trade history. Detailed trade uses granular payment data and captures monthly fluctuations in payment behavior and provides predictive improvement on scores.

Score system and model generation for viability rating

The ability to accurately assess risk depends on the availability of robust underlying data elements, so we have established a scoring system that considers the correlation between the depth of predictive data and future viability.

The exemplary result is a set of models consisting of four unique scorecards, each of which is driven by the depth of the predictive data elements, such as corporate identity data, which includes the size and industry of the business; Commercial payment transactions, which include the total amount of dollars owed three months ago; the available financial information attributes of the business, such as current ratios, and so on.

The viability score provides an ordering of, for example, 1 to 9 based on all four models combined. The portfolio comparison provides an ordering of, for example, 1 to 9 based on individual model sections. Providing two perspectives allows for a better understanding of the risks relative to all businesses and to those of the enterprise that are only within the same model segment. Having a model system allows for better separation of "good" and "bad" by focusing on unique groups. It also provides the most predictable score possible, which is optimized for available data. Therefore, the Survivability Rating uses a segmented scorecard to provide maximum risk identification to improve risk management decisions.

Table 1 below provides an estimate of the "bad" rate (eg, the rate of failure) based on untimely samples.

Each viability score has a "bad" rate that can be compared to the average. For example, Table 1 above shows that 1% of all companies with a score of 9 and 65% of the group are expected to close down, become inactive, or file for bankruptcy within the next 12 months. This means that the probability of a company with a viability score of 9 is roughly 5 times the average (65/14=5), and the probability of malpractice is 325 times that of a company with a score of 1 (65/0.2= 325).

The data depth indicator component acquires the known capabilities of the company and is used to create information on the viability score. The ability to measure viability fragments in terms of model accuracy and separation. However, in risk modeling, the model can have good accuracy, but there are many cases of poor performance in identifying good accounts and bad accounts. For successful risk analysis, it is important to distinguish between good and bad accounts. Therefore, a data depth indicator or score based on this particular modeling aspect is also used in generating the viability rating. There are many standard defined statistics, such as Kolmogorov-Smimoff, Gini index, divergence, ROC, etc., which acquire the separation ability of multivariate statistical models. The inventors have used the principal component analysis method to combine all of their statistics into An indicator or score. These scores are ultimately used to create weights for each dimension of the company that are used to calculate the viability score.

Weighting strategy for regression with multiple dependent variables

When using an OR condition to combine multiple binary dependent arrays into one dependent variable, for example, overall bad = bad 1 or bad 2 or bad 3. A bad definition with the highest NPL will dominate and mask other bad definitions. In the present application, the defect rate is 1 = 0.22%, the defect rate is 2 = 0.32%, and the defect rate is 3 = 0.12%. In the absence of a weighted regression model, it would be more accurate for bad 2 but less accurate for bad 1 or bad 3.

method:

To ensure that the regression model will work for all three bad definitions, the weighted bad rate and the number of weighted bads will be set such that for each of the three definitions, they will be equal in terms of counts and rates. Create a final weight set to ensure that the overall count and defect rate will be the same as the original data set to ensure proper interception and P statistic for unweighted samples. The following shows a series of tables showing the actual counts and weights used in the weighting scheme.

The first step is to increase the weighted bad count of bad 1 and bad 3 to the count of bad 2 . Poor 1 and Poor 2 and Poor 3 are mutually exclusive, but there is an overlap between Poor 2 and Poor 3. Due to the possibility that the account is both bad 2 and bad 3, a second weight must be applied, which will reduce the weight of the account that is bad 3 but not bad 3. Finally, a third weight is applied to bring the overall NPL and count back to the original unweighted data set (see Figure 6). FIG. 1A is a block diagram of a system 100 for the techniques disclosed herein. system 100 includes (a) computer 105, (b) data sources 145-1 and 145-2 through 145-N, collectively referred to as data source 145 and communicatively coupled to computer 105 via network 150.

Network 150 is a data communication network. The network 150 can be a private network or a public network, and can include any and all of: (a) a personal area network, for example, a room covered, (b) a regional network, for example, a building (c) campus area networks, for example, covering campuses, (d) metropolitan area networks, for example, covering cities, (e) wide area networks, for example, areas that are linked across metropolitan, regional, or national boundaries, or f) Internet. Communication via electronic signals and optical signals via the network 150 is performed.

Each of the sources 145 is an entity, organization or process that provides information about the business (ie, information). Examples of source 145 include business registration, phone book, account receivables invoice level payment information, and business inquiries about other businesses.

The computer 105 processes the data from the data source 145 and also processes the data designated as UTC data 129, account receivables data 130, detailed transaction data 135, and enterprise reference material 140 herein, and generates data designated as active signals ( ASD) 160 data and score 165 data.

Account receivables information 130 is account receivable information obtained from multiple companies that have supplied goods or services to other businesses or payments. The account receivable information 130 of the interested company is obtained from the goods or service provider of the company of interest. For example, assume that Company B is the goods or service provider of Company A. The account receivable due amount from Company A will be displayed on the book of Company B. In fact, there are likely to be many companies that supply goods or services to Company A, and therefore the account receivable information for Company A will be Includes accounts receivable information about A company from many companies.

Detailed transaction data 135 is additional information about the company of interest and may be derived from account receivables information 130. Examples of detailed transaction information 135 include the number of accounts that were overdue in the last six months, and the total amount outstanding.

Enterprise Reference 140 is a description of the company's information. For example, for a target company, the enterprise reference 140 will include the unique identifier of the underlying business, corporate information, financial statements, and traditional transactional materials. The unique identifier is the identifier that uniquely identifies the underlying business. The Global Data Encoding System (DUNS, Dunn) code can be used as such a unique identifier. Corporate information is information about the company, such as the number of employees, years of operation, and the industry in which the business is classified, for example, retail. The financial statements are financial information such as the quick ratio (ie (current assets - inventories) / current liabilities) and total liabilities. Traditional trading information is information such as the amount overdue for more than 30 days, the number of payment experiences overdue for more than 30 days, and the number of satisfactory payment experiences.

ASD 160 is information about the company from sources obtained from source 145. In general, ASD 160 indicates the level of processing activity for the subject company by other companies involved in the subject company.

Score 164 is a viability rating.

Detailed trade information 135, corporate reference 140, ASD 160, and score 165 are stored in one or more databases. One or more databases may be assembled as a single storage device or as a decentralized storage system having multiple independent storage devices. Although one or more databases are shown as being directly coupled to the computer 105 in the system 100, one or more databases may be located remotely to the computer 105 or coupled to the computer 105 via the network 150.

The computer 105 includes a user interface 110, a processor 115, and a memory 120 coupled to the processor 115. Although computer 105 is referred to herein as a stand-alone device, computer 105 is not limited thereto, but may be coupled to other devices (not shown) in a distributed processing system. The user interface 110 includes an input device, such as a keyboard or voice recognition subsystem, to enable a user to communicate information and command selections to the processor 115.

The user interface 110 also includes an output device such as a display or printer, or a speech synthesizer. A cursor control, such as a mouse, trackball or joystick, allows the user to manipulate the cursor on the display for communication of additional information and command selections to the processor 115.

The processor 115 is an electronic device that is composed of logic circuits that respond to instructions and execute instructions.

The memory 120 is a tangible computer readable storage device encoded with a computer program. In this regard, the memory 120 stores data and instructions that can be read and executed by the processor 115 to control the operation of the processor 115, that is, the code. The memory 120 can be implemented in a random access memory (RAM), a hard disk, a read only memory (ROM), or a combination of the above. One of the components of the memory 120 is the processing module 125.

The processing module 125 is a module of instructions, which are read by the processor 115 and control the processor 115 to perform the scoring of the enterprise, that is, by assigning the probability of default, wherein the probability of default is converted into The arrears score, that is, the score 165. The processing module 125 outputs the results to the user interface 110 and also directs the output to the remote device (not shown) via the network 150.

In this document, we describe the operation by the processing module 125 or its It is a process execution. However, the operations are actually performed by the computer 105, and more specifically by the processor 115.

In this document, the term "module" is used to indicate functional operations that can be implemented as separate components or as an integrated assembly of multiple accessory components. Therefore, the processing module 125 can be implemented as a single module or a plurality of modules that cooperate with each other. In addition, although the processing module 125 is described as being mounted on the memory 120, the processing module 125 can be implemented in a hardware (for example, an electronic circuit), a firmware, a software, or a combination of the above. Any of them.

Although the processing module 125 indicates that it has been loaded into the memory 120, the processing module 125 can be assembled on the storage device 199 for subsequent loading into the memory 120. The storage device 199 is a tangible computer readable storage medium having a processing module 125 stored thereon. Examples of the storage device 199 include a compact disc, a magnetic disk, a read-only memory, an optical storage medium, a hard disk or a memory unit composed of a plurality of parallel hard disks, and a universal serial bus (USB) thumb disk. Alternatively, the storage device 199 can be a random access memory, or other type of electronic storage device, and is located in the remote storage system and coupled to the computer 105 via the network 105.

In fact, data source 145, account receivables information 130, detailed trade information 135, and corporate reference material 140 will contain information representing a number of (e.g., millions) of data items. Therefore, in reality, human beings cannot handle the above information. On the contrary, the data needs to be processed by a computer such as computer 105.

FIG. 1B is a block diagram of the processing module 125. The processing module 125 includes a number of accessory modules, namely, an activity signal data (ASD) generator 205, an account receivables (A/R) process 210, a model generator 215, and a scoring process. 220. In short: (a) the ASD generator analyzes the data from data source 145 and generates ASD 160, as mentioned above, ASD 160 indicates the level of processing activity for the subject company by other companies involved in the subject company; (b) The A/R process 210 analyzes the account receivables information 130 from the supplier of the target enterprise and generates a weight indicating whether the underlying enterprise maintains a good reputation for debt payment or defaults on the payment of the debt; (c) model generation The processor 215 processes various enterprise profiles, ASDs 160, and weights from the A/R process 210, and generates models based on such data for scoring the enterprise; and (d) the scoring process 220 utilizes models from the model generator 215 Generate a score of 165.

Each of ASD generator 205, A/R process 210, model generator 215, and scoring process 220 is described in further detail below.

1C is a block diagram of an ASD generator 205 that analyzes data from a data source 145 and generates an ASD 160 as described above. The ASD generator 205 includes a matching process 305, a login process 310, and an aggregator 315.

As mentioned above, source 145 is an entity, organization or process that provides information about the business (ie, information). The format of the data is not specifically related to the operation of system 100, and for purposes of example, we will assume that the data is organized into records. Descriptor 301 is an example of such a record and contains material for describing various business aspects such as name, address, and phone number. In fact, descriptor 301 can include many such aspects.

The matching process 305 receives or obtains the descriptor 301 from the data source 145 and matches the descriptor 301 with the data in the enterprise reference 140.

As mentioned above, Enterprise Reference 140 is a description of the company's information. Enterprise reference materials 140 are organized into records. One such record, that is, record 340, is a representative example. Record 340 includes unique identifier 341, corporate information 342, financial statements 343, and traditional trade materials 344.

As used herein, matching means searching for data in a data storage device, for example, searching a database for a record that best matches a given query. Therefore, the matching process 305 searches the enterprise reference 140 for the material of the best matching descriptor 301.

The best match must be a correct match, so the matching process 305 also provides a credibility code after finding the match, which indicates that the correct level of confidence is matched. For example, the credibility code 5 may indicate that the match is absolutely correct, and the credibility code 1 may indicate that the match is correct with a relatively low certainty.

The matching process 305 generates a signal 306 after finding a match, the signal 306 comprising: (a) an identification of the source from which the data was received; (b) a time at which the match was made (including the date); and (c) a unique identifier 341; ) Trustworthy code.

Login process 310 receives signal 306 and logs it into the log, designated herein as metadata 320. Table 2 lists some exemplary metadata 320.

For example, the first row in Table 2 indicates that the matching process 305 generates a first signal, i.e., signal 1, which indicates that the matching process 305 at time t0 will be from the data source 145-2 descriptor 301 and the enterprise reference material. The data in 140 matches. The match indicates that the descriptor 301 is associated with the enterprise identified by the unique identifier 00000001 and the match has a confidence code of 2. In fact, meta-data 320 will contain many (eg, millions) rows of data.

Aggregator 315 aggregates the data from metadata 320 to generate ASD 160. More specifically, aggregator 315 considers metadata 320 that falls within a time period (i.e., period 312) and maintains a total number of signals for each unique identifier and has a greater than or equal to the confidence of threshold 313. The total number of matches for the code. Thus, for the underlying enterprise, ASD 160 includes a unique identifier 330, a number of signals 335, and a confidence code (CC) match 336. The number of signals 335 is the total number of signals for a particular unique identifier that was matched during period 312. The CC match 336 is the total number of matches having a confidence code greater than or equal to the threshold 313.

For example, referring to Table 2, assume that period 312 defines a time period from t0 to t4, and threshold 313 defines threshold 3. Table 3 lists the corresponding exemplary data for ASD 160.

Table 3 shows that during the period from t0 to t4, for the unique identifier 00000001, there are a total of three signals (see Table 1, signals 1, 3 and 4), and of the three signals, two are targeted for A match of the confidence codes greater than or equal to 3 (see Table 2, lines 3 and 4). Although not shown in Table 3, ASD 160 may include other information derived from signal 306, for example, an identification of a source 145 that provides a maximum number of confidence codes having a greater than or equal to threshold 313. match. In effect, cycle 312 will have a length that causes ASD generator 205 to aggregate a large number of events, for example, 12 months. Therefore, ASD 160 will include many (eg, millions) rows of material.

2 is a flow diagram of a scoring process for a viability rating (designated herein as method 200). The method 200 begins at step 202.

In step 202, computer 805 receives a company record to be scored from repository 840. In step 204, the company undergoes an entity matching process. In step 206, the unmatched record will get zero points.

In step 212, the data is appended to the records from all of the sources listed in Figure 1. In step 214, the company is inspected for availability and exclusion rules.

In step 216, data is recorded based on the record and the data is evaluated for model selection. Model selection depends on the availability of the data and its depth degree. For example, if a record has enough information from a financial statement, the record will be subject to an FN segment. If the record does not have significant trade activity, the record will be subject to the NT segment and the record will be evaluated based solely on firm statistical figures, smart engine signals, or other available material.

In step 218, the record will be subjected to a point assignment based on the value of the predictor from each data source. The predictor selection is based on which segment the record is suitable for.

During the scoring process (step 220), the points counted are summed to obtain the score and data depth size. Score the score for the first three components.

Next, in step 222, the company undergoes a set of queries to check for enterprise adjustments, including but not limited to special categories, such as high risk situations or failures. The rating is adjusted based on the particular category the company is classified into. There is a priority built into the adjustment rules to focus on the overall impact of known information on viability.

Based on the results from step 222, a final score and assignment of all categories of ratings is performed in step 224. If the company is not suitable for any adjustments, it has the same score as in step 220. If the company is suitable for adjusting the score, it carries the score from step 222. The demographic component of the rating is defined during the final scoring module (step 224).

Figure 3 is a description of the depth component of the data for the viability rating.

Figure 4 is a description of the company profile component of the viability rating.

Figure 5 is a depiction of the manner in which the four components of the scorecard are used to generate a viability rating. A company identity record is selected at module 502 for scoring. The data element is attached to the record. During the process in step 504 (model selection), the record is subjected to a series of queries to determine which model area the record should be subjected to. segment. In this particular case, the company has information from the financial statements that makes the company suitable to withstand the FN model (step 506). From steps 508 to 514, the viability and data depth points from each data source are summed to produce a viability score 516 and a data depth score 518.

In step 520, a demographic section is assigned. In the viability section 522, the viability score and asset portfolio comparison are calculated based on the score points from the viability score 516. The scores of the score points to the rating values are scored for the two viability components during step 522. In 524, the points of the data depth are mapped to the data depth rating. In step 526, the records are adjusted based on a particular category, including but not limited to a collapse or high risk situation. In this example, the record is not suitable for any adjustments and proceeds to step 528. In step 528, the final viability rating is presented or output to the user. The record projects the same scores as in steps 520, 522, and 524, respectively.

Figure 6 shows the value of the first component (viability score). For example, rating scales 1 through 9 are presented here. The cut-off value for each category is determined by the non-performing rate. The higher the value of the rating, the higher the risk of the company. Users will try to avoid "bad" businesses. Companies that are unlikely to survive and will not ultimately avoid good businesses. In the example shown, the overall NPL ratio was 19.9%. Companies that do not take advantage of this solution will eventually end up with 19.9% in their portfolio. By using the method of the present disclosure, the user can avoid segments 9 and 8 with higher defect rates and circumvent business with records from higher risk segments.

The use of survivability ratings in this disclosure is numerous, ranging from risk assessment to supply chain analysis to market use for pre-screening or improved target market selection.

As an example, large banks are trying to expand their loan asset group Hehe. Using the survivability rating of this disclosure, the bank found that this viability rating identified multiple segments with a response rate four (4) times higher than the conventional rating system.

While we have shown and described several embodiments in accordance with the present disclosure, it is to be understood that the embodiments are subject to numerous modifications and modifications are apparent to those skilled in the art. Therefore, we do not intend to be limited to the details shown and described, but are intended to show all changes and modifications within the scope of the appended claims.

140‧‧‧Corporate References

160‧‧‧Active signal data

165‧‧‧ score

200‧‧‧ method

202, 212, 214, 216, 218, 220 ‧ ‧ steps

210‧‧‧A/R processing/steps

Claims (19)

A method for determining future commercial viability of an entity, the method comprising: (a) determining a future commercial viability of the entity using a first predictive modelling by identifying data a pattern and associated with the predicted attributes to derive, thereby generating a viability score; (b) using a second predictive modelling to generate a relative ranking of the entity relative to one of its peer groups, thereby generating a comparative viability score (c) measuring the depth of the data to quantify the level of understanding of the entity and the degree of trust in the viability score and the comparative viability score, thereby generating a data depth indicator; (d) assigning a company profile, which borrows Defining the entity by segmentation and grouping the entity with other similar entities; and (e) outputting a multi-dimensional including the viability score, the comparison viability score, the data depth indicator, and the company profile Vitality rating. The method of claim 1, wherein the company profile defines the entity in at least one selected from the group consisting of: and classifies the entity and other similar entities into groups: size, years of operation, complete financial statements Availability and history of commercial trade. The method of claim 1, wherein the viability score is a predictive rating on a viability score scale. The method of claim 3, wherein the viability score scale is between about 1 and about Within the range between 9 and 1 are the lowest probability that one entity will close or become inactive over a period of time, and the 9th will collapse or become inactive. The method of claim 1, wherein the comparative viability score is a predictive rating on a comparative survivability score scale. The method of claim 5, wherein the comparative viability score scale is in a range between about 1 and 9, wherein the ratio of 1 is lower than that of other companies in a period of time or becomes inactive, and The highest probability that the 9 series will fail or become inactive. The method of claim 1, wherein the data depth indicator is based on a descriptive rating of one of a data depth identifier scale. The method of claim 7, wherein the data depth indicator scale is in a range between approximately A and M. The method of claim 8, wherein the A to G are assigned on a "class report card" scale, wherein A is assigned to the enterprise with the highest level of forecast data selected from the group consisting of: : Complete corporate statistical structure, extensive commercial trade activities, integrated financial attributes and their mixtures; and assign G to a company with the lowest level of forecast data. The method of claim 9, wherein the predicted data is basic identity data. The method of claim 8, wherein the H to M is a special category that is prioritized over the A to G rating, giving the user further insight when confirming that an enterprise has encountered one of a set of predefined risk conditions force. The method of claim 1, wherein the company profile is based on a descriptive rating of a company profile. The method of claim 12, wherein the company profile is in a range between about A and Z. The method of claim 13, wherein A has the largest, longest-established enterprise with complete reported comprehensive data, and X is the smallest, youngest enterprise with basic corporate identity data. A computer readable storage medium containing executable computer program instructions that, when executed, cause a processing system to perform a method for determining future commercial viability of an entity, the method comprising: (a) using one The first prediction model is used to determine the future commercial viability of the entity, and the first prediction model is derived by identifying the type of the data and correlating with the predicted attribute to generate a viability score; (b) using one The second prediction modelling produces a relative ranking of the entity relative to one of its peer groups, thereby generating a comparative survivability score; (c) measuring the data depth to quantify the level of knowledge of the entity and the survivability score and Comparing the degree of trust of the survivability score, thereby generating a data depth indicator; (d) assigning a company profile that defines the entity by segmentation and grouping the entity with other similar entities; and (e Outputting a multi-dimensional viability rating including the viability score, the comparative viability score, the data depth indicator, and the company profile. A computer system for determining the future commercial viability of an entity, the system comprising: a database containing active signal data; An activity signal generator that aggregates the activity signal data using a plurality of data sources from a plurality of businesses doing business with an entity of interest; and a model generator based on a statistic The model produces a viability score, where a dependent variable performance is derived from independent variables using statistical probabilities, which are created from multiple sources. The system of claim 16, wherein the processor performs the following steps stored in the memory; the steps comprising: (a) determining a future commercial viability of the entity using a first predictive modeling, the first Predictive modeling is derived by identifying the type of data and correlating with predictive attributes to produce a viability score; (b) using a second predictive modelling to generate a relative ranking of the entity relative to one of its peer groups And generating a comparative survivability score; (c) measuring the depth of the data to quantify the degree of knowledge of the entity and the degree of trust in the viability score and the comparative viability score, thereby generating a data depth indicator; Assigning a company profile by segmenting the entity and grouping the entity with other similar entities; and (e) outputting one including the viability score, the comparison viability score, the data depth indication And the multi-dimensional viability rating of the company profile. The system of claim 16, wherein the activity signal generator comprises: a matching process that generates a signal when a match is found; a login process that receives the signal and inputs the signal to the source And an aggregator that aggregates data from the metadata to generate the active signal data. The system of claim 18, wherein the signal comprises at least one selected from the group consisting of: identification of the source, data received from the source; time to form the match; unique identifier; Credibility code.