Accelerated Intelligent Enterprise with Timely Vendor Spend Analysis

Abstract

Disclosed are intelligent enterprise systems and methods that include an input / output (IO) interface configured to receive information related to accounting information, a processor interconnected with a memory configured to store the information received from the IO interface and analyze the information to provide timely vendor spend analysis, and a display device configured to display the provided timely vendor spend analysis.

Description

Cross - reference to related applications

[0001] This application claims the benefit of U.S. Patent Application No. 62 / 987,623, entitled "ACCELERATED INTELLIGENT ENTERPRISE INCLUDING TIMELY VENDOR SPEND ANALYTICS", filed on March 10, 2020, and is incorporated by reference in its entirety as if fully set forth herein.

Technical Field

[0002] The present invention is directed to enterprise systems, and more particularly, to intelligent enterprise systems including timely vendor spend analytics.

Summary of the Invention

[0003] Disclosed are an intelligent enterprise system and method. The system and method include an input / output (IO) interface configured to receive information related to accounting information, a processor interconnected with a memory and configured to store the received information from the IO interface and analyze the information to provide timely vendor spend analytics, and a display device configured to display the provided timely vendor spend analytics.

[0004] The system and method may include received information including category - based spend. The system and method may include received information including vendor - based spend. The system and method may include received information including spend over a period such as a selected fiscal year.

[0005] The system and method may include received information processed by a processor to find exact matches using nearest neighbor trend scoring. The system and method may include received information processed by a processor to identify matches using a fuzzy matching algorithm. The system and method may include received information processed by a processor to identify matches using at least two fuzzy matching algorithms. The system and method may include matching based on a cutoff score. The system and method may include parent company information identified for at least one vendor.

[0006] The system and method may include performing text tokenization. The system and method may include OEM learning based on received information. The system and method may include determining taxonomy categories based on received information.

[0007] The system and method may include utilization of at least one of a hybrid model combining a support vector machine (SVM), a random forest (RF), a neural network (NN), and an expert system (ES).

[0008] The system and method may include that a display device is configured to select aspects of vendor actual expenditures, peer comparison, market considerations, and regulatory considerations.

[0009] A more detailed understanding can be obtained from the following description given by way of example in conjunction with the accompanying drawings, in which like reference numerals in the drawings indicate like elements.

Brief Description of the Drawings

[0010]

Figure 1

[0011]

Figure 2

[0012]

Figure 3

[0013]

Figure 4

[0014]

Figure 5

[0015]

Figure 6

[0016]

Figure 7

[0017]

Figure 8

Figure 9

[0018]

Figure 10

[0019]

Figure 11

[0020]

Figure 12

[0021]

Figure 13

Figure 14

[0022]

Figure 15

[0023]

Figure 16

DETAILED DESCRIPTION OF THE INVENTION

[0024] The present invention is directed to an intelligent enterprise including vendor expense analysis. The present invention can accelerate enterprise intelligence using artificial intelligence for analysis. The enterprise provides timely, transparent, and regulatory considerations. Users can make wiser and faster decisions with reduced risks through enterprise collaboration and the use of regulatory intelligence. The enterprise may be used with an operating financial and vendor management solution. The enterprise utilizes a non-dependent overlay that is a SaaS application for classifying and visualizing vendor expenses in conjunction with all financial systems.

[0025] As used hereinafter, the term "processor" includes, but is not limited to, single or multi-core general-purpose processors, dedicated processors, conventional processors, digital signal processors (DSPs), multiple microprocessors, one or more microprocessors associated with DSP cores, controllers, microcontrollers, one or more application-specific integrated circuits (ASICs), one or more field-programmable gate array (FPGA) circuits, other types of integrated circuits (ICs), system-on-chip (SOC) and / or state machines. As used hereinafter, the term "computer-readable storage medium" includes, but is not limited to, semiconductor memory devices such as registers, cache memory, read-only memory (ROM), dynamic random access memory (D-RAM), static RAM (S-RAM) or other RAM, magnetic media such as flash memory, hard disks, magneto-optical media, optical media such as CD-ROM, digital versatile disks (DVDs) or Blu-ray disks (BD), other volatile or non-volatile memories or other types of devices for electronic data storage. As used hereinafter, the term "memory device" is a device configurable to read and / or write data to one or more computer-readable storage media.

[0026] When referring to the following, the term "display device" includes, but is not limited to, a monitor or television display, a plasma display, a liquid crystal display (LCD), or a display based on technologies such as front or rear projection, light emitting diodes (LEDs), organic light emitting diodes (OLEDs), or digital light processing (DLP). When referring to the following, the term "input device" includes, but is not limited to, a keyboard, a mouse, a trackball, a scanner, a touch screen, a touch pad, a stylus pad, and / or other devices that generate an electronic signal based on an interaction with a user person. The input device may operate using technologies such as Bluetooth, Universal Serial Bus (USB), PS / 2, or other technologies for data transmission.

[0027] FIG. 1 shows an example computing system 100 for determining vendor spend analysis. In a vendor spend analysis of a service provider, it is desirable Performance to be easily comparable with both a benchmark and other service providers Performance in the same industry or organization. The service provider Performance is shown. As will be described in more detail below, the vendor spend analysis is Performance a weighted sum based on a benchmark value and a plurality of weighting factors. The vendor spend analysis may be calculated Enterprise's by combining information about spend data from a financial system , business-related metadata and For example, vendor metadata such as risk scores, diversity assessments, and other data fields may be used to contextualize the spend data. By collecting these data points across multiple companies, a peer comparison of the collected information may be provided.

[0028] One example of a computing system 100 includes a processor 165, a memory device 170, and a communication bus 180 for data communication between components. The computing system 100 optionally includes a plurality of components, including a graphics subsystem (not shown), a network interface (not shown), an input interface 160, and a printer driver (not shown). The graphics subsystem may provide data to and / or drive a display device 140. The network interface provides a wired or wireless network connection through which the computing system 100 can connect to a network-connected printer or the Internet. The input interface 160 may be, for example, a wireless connection or a data port capable of receiving data from one or more input devices. The printer driver provides data to and / or drives a local printer.

[0029] The computing system 100 is a service provider Performance evaluates data, Performance generates graphical output data for displaying the results of a determination of vendor spend analysis based on the data, and includes a subsystem for analyzing the vendor spend analysis and related data. The subsystem is PerformanceIt includes a data module, an aggregation module, a quantification module, a vendor expenditure analysis module, a display module, and a result analysis module. The modules may be implemented as software modules, dedicated processor elements, or combinations thereof. Suitable software modules include, by way of example, executable programs, functions, method calls, procedures, routines or subroutines, sets of processor-executable instructions, scripts or macros, objects or data structures. Some or all of the modules may also be implemented as modules or subroutines in a spreadsheet program such as Microsoft Excel (registered trademark), a database program, or other types of programs. The data used and / or generated by the modules may be stored in the memory device 170. The processor 165 can be configured to operate and control the operation of the modules, and can also be configured to control the communication between the modules and other components of the computing system 100, such as the printer driver, the graphics subsystem, the memory device 170, the network interface, and the input interface 160.

[0030] Generally, the accounting system 120 is used to summarize, analyze, and report on a company's financial operations. The accounting system 120 may include, as described below, a ledger-expense-credit-revenue system 128, a debt system 126, assets 124, and a budget 122. This The accounting system 100 te performs vendor expenditure analysis Decision through information such as the ledger-expense-credit-revenue system 128, the debt system 126, assets 124, and the budget 122. The information in the accounting system 100 includes data supplied from the billing system, purchase orders, accounts payable, invoice details, a general ledger chart of accounts, and the business hierarchy. The data from these information sources are cross-referenced by the system 100 so that the data are organized, classified, and enriched.

[0031] Generally, the ledger of the ledger-expense-credit-revenue system 128 may have a general ledger. The general ledger represents a record-keeping system for a company's financial data, including records of debit and credit accounts verified by a trial balance. The general ledger provides records of each financial transaction conducted in the operation of a business company. The general ledger holds the account information necessary to prepare a company's financial statements, and the transaction data is separated by type into accounts for assets, liabilities, equity, revenue, and expenses.

[0032] The expenses of the ledger-expense-credit-revenue system 128 are records within the ledger that represent the occurrence of payments or liabilities in exchange for goods or services. The documentary evidence generated by the expenses is a sales receipt or invoice.

[0033] Debits and credits are entered into the ledger accounts to record changes in value due to business transactions. An entry of a debit in an account represents the movement of value into that account, and an entry of a credit represents the movement out of the account. For example, a renter who pays rent to a landlord may enter a debit in the rent expense account corresponding to the landlord, and the landlord may enter a credit in the accounts receivable account corresponding to the renter. In all transactions, both a debit entry and a credit entry occur for each party involved, and the total of the debit entries and the total of the credit entries for each party for the same transaction are equal.

[0034] The revenue of the ledger-expense-credit-revenue system 128 is the income that a business obtains from its normal business activities, usually the sale of goods or services to customers. Revenue is also called sales or sales revenue.

[0035] Accounts payable (AP) 126 is an account in the general ledger that represents a company's obligation to repay short-term debts to creditors or suppliers. AP126 includes the accounting for payments that a company should make to suppliers and other creditors. AP126 is the amount still unpaid that should be paid to the vendor or supplier for goods or services received. The total of all outstanding balances to be paid to vendors is shown in the company's balance sheet as the accounts payable balance.

[0036] Asset 124 is all resources owned by an enterprise. Tangible or intangible things that can generate value through ownership or control and are held by an enterprise to generate beneficial economic value are assets. Simply put, assets represent the value of ownership that can be converted into cash.

[0037] Budget 122 represents a financial plan that includes both financial information and non - financial information. Its most obvious features are the revenue estimate, the selling expenses for a given organization, and the costs. The budget can also include non - financial information such as how many employees are considered necessary. Budget 122 is a document of prediction, but enterprises also use it as a financial control tool. Financial control is a tool for monitoring the activities of one's own business.

[0038] The vendor expenditure analysis module may be included within data system 150 and / or accounting system 120. In vendor expenditure analysis, a weighting coefficient is determined for each quantitative Performance value, each quantitative Performance value is multiplied by the corresponding weighting coefficient, and the sum of the products of the multiplied quantitative Performance value and the weighting coefficient may be used to judge the vendor expenditure analysis. The weighting coefficient may be based on user input received through the input interface. The weighting coefficient may be adjusted so that the vendor expenditure analysis focuses on the aspects of the criteria that the user deems most important and / or does not emphasize the criteria that the user deems unimportant. In an example of a use case, the user may set a Performance coefficient to indicate that a particular Performance criterion is extremely important. At a later point in time, the user may adjust the weighting coefficient to reflect that other Performance criteria have become more important. Performance criteria have become more important.

[0039] The display module may receive the data generated by the quantification module and the vendor expenditure analysis module, and generate data for displaying the received data. The display data may be communicated by the display module to the printer driver and the printer, and the printer may create a printed version of the display data. The display data may be communicated by the display module to the graphics subsystem for the display device 140, and the display device 140 may render the display data for the user. A visual depiction example of the display data generated by the display module and printable by the printer or renderable by the display device 140 will be described in detail below.

[0040] The display data generated by the display module can take various forms according to various implementations of the display module and the computing system 100. For example, the generated display may represent the content of an image, one or more pages within a document, or other visual artifacts. The generated display data may be compressed or uncompressed image data conforming to formats such as Joint Photographic Experts Group (JPEG), Tagged Image File Format (TIFF), Scalable Vector Graphics (SVG), bitmap, or other formats. The generated display data may be data or instructions interpretable by a window manager, a desktop manager, a graphical user interface, a printer driver, a graphics driver, a word processor, a document display or editing program, or other software or hardware components related to the implementation of the user interface and / or graphics.

[0041] The result analysis module may receive the data generated by the vendor expenditure analysis module and determine whether the parameters used by the accounting system 120 and the vendor expenditure analysis module should be changed. For example, the result analysis module may compare the vendor expenditure analysis values generated for multiple vendors. If most of the vendors have vendor expenditure analysis values exceeding the threshold, the result analysis module may adjust the parameters used by other modules so that the generated vendor expenditure analysis values decrease. Alternatively, if most of the vendors have vendor expenditure analysis values below the threshold, the result analysis module may adjust the parameters used by other modules so that the generated vendor expenditure analysis values increase. The parameters that can be changed by the result analysis module include Performance value criteria, Performance value sub-criteria, and the weighting factors used by other modules.

[0042] Also, the accounting system 120 may perform e-commerce or banking functions based on the data received from the quantification module and / or the vendor expenditure analysis module. Data indicating past payments to service providers may be made available for use by the result analysis module in the memory device and / or via the network interface. Payments to service providers may be conditional on subsequent vendor expenditure analysis, Performance benchmark values, and / or Performance sub-benchmark values, and Performance rules associating them with rewards may be stored in the memory device. The accounting system 120 may determine that a debit or credit should occur for the account of the service provider based on the reward rules and the data received from the quantification module and / or the vendor expenditure analysis module. Then, the accounting system 120 may initiate an electronic transaction to generate the determined debit or credit. For example, if the result analysis module determines that the vendor expenditure analysis or service PerformanceIf it is determined that the standard or sub - standard has been exceeded, the result analysis module may start paying a bonus to the service provider. If the result analysis module determines that it has overpaid the service provider for the services performed, the result analysis module may start a debit to the service provider's account. The result analysis module may initiate a credit or debit using technologies such as Electronic Funds Transfer (EFT, electronic funds movement) or other appropriate services or protocols via a network interface.

[0043] Also, the accounting system 120 may analyze the data received from the quantification module and / or the vendor expenditure analysis module based on a trigger, and notify the user when the trigger - targeted event occurs. For example, the trigger may be set based on whether a service provider exceeds a required vendor expenditure analysis, one or more Performance reference values, or one or more Performance sub - reference values. Alternatively, the trigger may be set based on whether a plurality of service providers have one or more Performance reference values, or one or more Performance sub - reference values for the required vendor expenditure analysis. If a trigger condition exists, the result analysis module may generate a notification email indicating that the trigger condition has been met for sending to the user. Alternatively, the accounting system 120 may communicate with the display module to generate display data indicating that the trigger condition has been met. Then, the corresponding notification may be displayed on the display device 140 for the user. The user may use the notification, for example, as a basis for renegotiating the terms of the SLA with one or more service providers.

[0044] The computing system 100 may be implemented, for example, in a computer, a system-on-chip, or other computing or data processing device or apparatus. The processor 165 may be configured to execute instructions that specify the functionality of the modules as described above with reference to FIG. 1. The instructions may be stored in the memory device 170 and / or one or more additional computer-readable storage media (not depicted).

[0045] FIG. 2 shows an exemplary screenshot 200 of the system 100 of FIG. 1 that conveys vendor expense analysis by providing expense 210 by category. In screenshot 200, it is shown that the company used $149.9 million in fiscal year 2019 220. And this expense data is separated into specific expense categories and depicted using the color and size of the display portion to indicate the direction and magnitude of the amount used for each category. As shown in screenshot 200, the application 230 has the largest expense, corresponding to the largest depicted box and corresponding to an expense amount of $66.9 million. The big data 240 is the second largest expense and is depicted using the second largest box representing an expense amount of $27.1 million. The computing 250 is the third largest expense representing $26.5 million and is somewhat smaller in size than the representation of the big data as it represents a somewhat smaller expense. Also shown are expenses for other categories including non-IT 260, platform 270, end user 280, delivery 285, security 290, and IT management 295. Each of the respective boxes is sized according to the amount of expense corresponding to the category.

[0046] As an example, the entire display may be utilized to represent the overall expenses. From there, the size of each category may be made to follow a percentage representing the proportion that the category's expense occupies of the total expenses.

[0047] In addition to the size of the box being described, the color of the box may also be used to provide additional information. Such information may include trends in spending on categories.

[0048] Figure 3 shows an exemplary screenshot 300 of the system 100 of FIG. 1 that conveys a vendor spending analysis by providing spending 310 by vendor. Similar to the depiction in FIG. 2, the screenshot 300 of FIG. 3 again shows that the company used $149.9 million in the 2019 fiscal year 320. And this spending data is separated by specific vendor and depicted using the color and size of the display portion to indicate the direction and magnitude of the amount of money used by each vendor. As shown in screenshot 300, the first vendor 330 has the largest spending, corresponding to the largest depicted box and corresponding to a usage amount of $36.3 million. The second vendor 340 has the second largest spending and is depicted using the second largest box representing a usage amount of $22.6 million. The third vendor 350 has the third largest spending representing $18.2 million and is slightly smaller in size than the representation of the second vendor 340 since it represents a slightly smaller spending. As further spending, the spending of other vendors including the fourth 360 through vendor x is also shown. Each of the boxes is sized according to the amount of spending corresponding to the vendor.

[0049] As an example, the entire display may be utilized to represent the overall spending. From there, the size of each vendor may be made to follow a percentage representing the proportion of the vendor's spending to the overall spending.

[0050] In addition to the size of the box being described, the color of the box may also be used to provide additional information. Such information may include trends in spending on vendors.

[0051] Figure 4 shows a vendor roll-up flow 400 corresponding to the present system 100 of FIG. 1. In the vendor roll-up shown in FIG. 4, vendor data is extracted from the database 41 of the system 100 of FIG. 1. With this extracted vendor data, for example, it becomes possible to illustrate to users such as CIO, CMO, and CFO how much a company has spent in total on multiple vendors rolled up into one company through corporate acquisitions such as mergers and acquisitions. The vendor roll-up may be referred to as parentage.

[0052] In a vendor roll-up, the manufacturers of purchased items are all identified in the OEM disclosure, and furthermore, links to OEM owners are made possible. In an automatic and real-time or near real-time manner, a purchasing company may be notified that some vendors have been acquired by other companies. This feature also enables notifications regarding potentially redundant, wasteful, and inadequate transaction negotiations.

[0053] For example, a large financial institution spends $30 million a year on VMWare, $5 million on Pivotal, $25 million on EMC, and $20 million on Dell. The purchasing staff in the company's procurement department may not be aware that Dell has acquired VMWare, Pivotal, and EMC. Therefore, instead of negotiating each contract separately at a lower amount, the large financial institution may consider a lump-sum expenditure amount of $70 million for Dell, which is much higher than the individual amounts. With this higher expenditure amount, a perspective of negotiating better pricing can be obtained in negotiations with each of the commonly owned vendors. Substantial savings that can reach tens of millions of dollars can be achieved.

[0054] Identifying actual companies requires normalization 420 of supplier names and OEM company names. In the method of FIG. 4, when importing company names, it is allowed to shorten them to initials, synonyms, doing business as (DBA), or also known as (AKA). As a further improvement in the method, those that are only located in different states or countries may be regarded as the same company. Also, to create the normalized name, the name may be stemmed in part or in whole. This normalized name is retained in the database of FIG. 1 and reused by being associated with the appropriate company.

[0055] Also, multi-stage fuzzy matching algorithms 4301, 4302 are utilized to further perform normalization. The algorithms use a statistically tested cut-off score to produce accurate matching results. FIG. 4 shows a detailed diagram of the process flow.

[0056] Specifically, FIG. 4 shows the organizational database of FIG. 1 that provides the input associated with the new companies and vendors to be searched to a first step 420 of identifying the nearest neighbors with a propensity score matching algorithm to find a perfect match. The output of this test is provided to an algorithm 450 that determines the parent company of the vendor or company. The algorithm also outputs a perfect match to the second algorithms 4301, 4302 that use fuzzy matching as described above in this specification. The output of the fuzzy matching 4301 is provided to the second fuzzy matching algorithm 4302, and item 440 is filtered by a predetermined cut-off score. And these values are input to an algorithm 450 that determines the parent company of the vendor or company. And the result is fed back to the organizational database 470 to output a final data set 460 for the vendor OEM roll-up.

[0057] The system intelligently determines and reveals the vendor parent-child relationships resulting from mergers and acquisitions. This solution can reveal important relational data in a shorter time than any existing process or solution, because current methods fail to normalize company names, which leads to inaccurate associations even if an association can be made. The system improves the parent-child information to include all known past parent organizations and the dates when the parentage changed, enabling the ability to analyze spending at a specific time, which is particularly important when considering our benchmarks and trends.

[0058] Similar to what was described above for Figure 4, accounts payable, invoice, and purchase order data are used to determine a company's spending on vendors at the OEM and vendor parent levels. By using artificial intelligence and machine learning in the vendor taxonomy, the original OEM is determined regardless of who the goods or services were purchased from or whether OEM data is included in the customer data. The combination of artificial intelligence and machine learning results in "reseller spending visibility," which reveals and visualizes the actual amount of a company's spending on original equipment manufacturers ("OEMs") that is unknown in most systems due to purchases from multiple and numerous resellers or through them.

[0059] For example, a company is spending over $2 million per month ($24 million per year) on Cisco products, but this is not apparent from any of the company's financial records. This is because purchase orders, requisitions, and invoices show that Cisco products are purchased through a number of VARs (value-added resellers) and resellers. Large organizations can negotiate better terms by presenting where a vendor sits on the company's vendor list. In this example, if Cisco is one of the top five vendors, leverage could be applied based on this relationship, for example. Such information is not revealed in existing financial systems when OEMs require an organization to purchase their products through VARs or resellers.

[0060] This system performs vendor consolidation similar to that described above. This occurs not only at the OEM level but also integrates "non-OEM" vendor spend. The need for vendor consolidation arises from multiple vendor instances within the raw data (inconsistent counterparty vendor tables) and corporate affiliations that exist outside the raw data. Expenditure amounts may be classified into standardized technology business management categories to promote transparency in showing how the money paid to vendors is distributed or used within the organization.

[0061] In the database of FIG. 1, a core OEM hierarchy table is constructed. The table links to itself to form a parent-child hierarchy. Expenditure data from customers is imported and matched with potential names by machine learning that can be specific to the supplier name (vendor in the purchase record), OEM name, manufacturing part number, model number, and customer data. This approach uses extended NLP (Natural Language Processing) to process the expenditure description. Then, the result of the expenditure description is passed to a deep learning model that includes an inference engine with heuristics that provide speed and scale. Also, a linear support vector machine and a random forest process the final output. The final output 460 is used to determine whether the input OEM name matches an existing OEM name. If system 100 does not have sufficient certainty about the matching, the combination of the purchase data and the input OEM name is passed through a neural network. If there is still not a sufficiently high certainty about the matching, the input OEM name is passed to a human curator for analysis. The human curator attempts to match it with an existing OEM name or indicates that this is a new OEM. The system creates a new OEM if necessary. When a match is indicated either manually or by the neural network, subsequent manual curation is not required because the input OEM name is added as an alias for the existing OEM. Once the expenditure data is analyzed and matched, the usage amount can be aggregated by the parent OEM suitable for the OEM ownership hierarchy at the time of expenditure.

[0062] The OEM hierarchy table is kept up-to-date by a combination of manual curation and the automatic import of organizational hierarchy data from companies such as LexisNexis, Dun & Bradstreet, Moody's, and open ownership. The OEM hierarchy is also tracked by date ranges because the corporate ownership of an organization changes over time.

[0063] Most companies fail to accurately report vendor and OEM expenses. This is because accounts payable and PO systems are not designed to report at this level of specificity, and manual data entry when creating purchase orders, such as product information, is often not systematized. Additionally, financial reporting systems prioritize financial reporting. The accounts payable system does not necessarily need to capture detailed ledger items as part of the payment record for product information. Typically, the financial reporting system does not include vendor details for expenses recorded and processed through the fixed asset system. Even in financial system reports, data may be aggregated based on general ledger accounts, but generally, there is no retention of sufficient vendor details to provide accurate vendor reporting. This system 10 intelligently determines OEMs through a combination of technologies and achieves a higher matching rate than any existing process or solution in a short time.

[0064] Accurately aggregating vendor expenses is a huge task. In large enterprises, vendors often overlap in accounts payable and purchase order systems, which is due to user biases such as remittances to multiple addresses, regional instances of a given vendor worldwide, for example, the use of inconsistent abbreviations in the entered vendor names, and the maintenance of generally inconsistent vendor tables. This system normalizes and organizes inconsistent vendor names by "compressing the names". For example, in the accounts payable vendor table in Figure 1, IBM (I.B.M.) can be seen in multiple forms. These may include, for example, IBM (I.B.M.), I.B.M. (I.B.M.), International Business Machine (International Business Machine), International Business Machine S.A (International Business Machine S.A.). Training the system to recognize each name as the same organization is essential for efficient and accurate vendor aggregation.

[0065] When processing a transaction, the first items to be resolved are the supplier's name and the manufacturer's name. These may vary in spelling accuracy, use of kana, and abbreviation. For example, Dell (デル), Dell Technologies, Inc. (デル テクノロジーズ、 インク.), Dell Corp ltd (デル コーポ エルティーディー) should all refer to Dell (デル). For example, HP (エイチピー) and Hewlett Packard (ヒューレット パッカード) are the same, but HPS (エイチピーエス) is different. The system 100 matches the input vendor name with the existing hierarchy by interpreting and normalizing various variations of the company name. Also, the system 100 simplifies the dataset through a visualization engine.

[0066] Figure 5 shows a flow 500 for searching for OEMs from products in the database of FIG. 1. The item description is used to train the model to recognize words for classifying OEMs. Manufacturing companies use turns of phrase created to differentiate their products. This turn of phrase is not necessarily captured at the invoice line item level, but is often captured at the invoice line item level. In topic modeling, the concept of word vectors may be used to search for topics of interest, and in this system, it is used to search for OEMs from product descriptions. This approach requires a large corpus of data from OEMs to be associated with a high degree of certainty.

[0067] The implemented flow 500 takes an incremental approach to learning OEMs. The OEMs are verified and the model is updated to improve the next training iteration.

[0068] Flow 500 may further utilize online learning by forming a strong foundation that gradually improves, taking into account the fact that the words describing the category can change over time. Treat each word as a random variable and make multiple associations the conditional probability that those words are the given OEM. The set of features that are the words used can change over time. And this set of features is defined by Equation 1.

Number

Number

[0069] While other organizations use catalogs of products manufactured by OEMs, this system uses a combination of artificial intelligence and machine learning, along with purchase item data, reseller information, and expense classifications, to determine the OEM based only on product descriptions. Additionally, this system differentiates the use of product-level information with a visualization engine.

[0070] Specifically, FIG. 5 shows a flow in which, at step 510, data is input in the form of PDF (Portable Document Format), CSV (Comma-Separated Values), or other file formats. At step 520, the text description of the invoice PO line items is determined. A random forest method may be used to learn the categories, and variable importance may be used to prune the dictionary of terms used. Also, at step 530, the system may convert the text to terms and then remove N-grams that do not support class separation using term frequency and inverse document frequency.

[0071] At step 540, stemming may be performed on the text description. This may include removing stop words and creating N-grams of length 1 to 3 words across the entire word while maintaining the order. The enriched data from step 530 may be further fed into the stemming.

[0072] At step 550, the system may learn which N-grams match the categories. From there, at step 570, the OEM may be learned, and at step 590, a taxonomy category may be learned to identify the OEM from the description.

[0073] In addition to providing transparency about who a company is spending money on, an equally important data point is providing transparency about what a company is purchasing. By using a predefined and recognized taxonomy, combined with the ability to surface data sets through the taxonomy, the system is able to provide a standardized context for spend aggregation. The standardized taxonomy enables analyses such as benchmarking and anonymous peer-to-peer comparisons. A custom taxonomy via the customer portal may be configured to enable customers to create further data visualizations.

[0074] For companies, it is fundamental to define a given number of levels for their taxonomy and render them as a single row within the database. This system can use any number of levels to accommodate all vendors and provides maximum flexibility. This system may be used to encompass both IT and non-IT spending.

[0075] The taxonomic categories to be set may be determined first. Then, the model has to learn to automatically classify the expenditure account items into appropriate categories. This automatic classification is achieved by using a hybrid model that combines a support vector machine (SVM), a random forest (RF), a neural network (NN), and an expert system (ES). The random forest may be used to detect n-grams correlated with the categories. Through this process, many variables are reduced from 6,000 to between 200 and 500. These variables may be used to construct the SVM and the NN. In the SVM and the NN, the results are adjusted to show the probability that an item actually belongs to a class. This probability is used to determine which items need to be manually reviewed for appropriate class assignment. Using both the SVM and the NN is equivalent to obtaining the consensus of experts by having two experts confirm and verifying that an item belongs to a specific class. If this method produces conflicting results, the classes are selected for that item or sent to a person for review based on the reliability of the method and the difference in that reliability.

[0076] ES, which is an expert system, is for items that have excellent inference rules for determining the correct class label. This is a preliminary process until the SVM, RF, and NN obtain sufficient capacity to learn the categories and perform automatic classification. The RF has a high execution time cost but highly agrees with the way of inference in which a professional human determines the class label.

[0077] This approach utilizes both a hybrid approach and an approach in which data is prepared through learning to succeed in taxonomy. It is essential to have a trade-off with the expert system rules to incorporate inference rules from professional humans and supply the results to the engine for training. The initial data volume is not large, and means for obtaining excellent labeled data are required. The expert system provides this bridge.

[0078] FIG. 6 shows a screen depiction 600 of the enterprise system 100 of FIG. 1, which shows a navigation window for selecting vendor actual expenses 610, peer comparison 620, market consideration 630, and regulatory consideration aspect 640. This screen may be used to create and utilize aggregated data from the database of FIG. 1. The system launches an actual expense analysis by selecting an actual expense navigation selection button.

[0079] FIG. 7 shows a screen depiction 700 of the enterprise system 100 of FIG. 1, which shows a landing page for the category expenses 610 described above in this specification. This uses the taxonomy described above in FIG. 2, including the application 230, big data 240, calculation 250, platform 270, end user 280, delivery 285, security and compliance 290, IT management 295, and non-IT (not shown). The presented IT patterns are provided, and in this example, the annual expenses 710 are provided. For example, each of the variables may be changed to change the presentation by using other colors for the trends or other display time frames.

[0080] Figures 8 and 9 show a screen depiction 800 of the enterprise system 100 of FIG. 1, showing vendor expenditures after selecting the vendor expenditure 610 in the window 700 of FIG. 7. FIG. 8 represents a screen depiction 800 that is partially deformed from the screen 700 of FIG. 7, and FIG. 9 represents a screen depiction 800 that is more completely deformed from the screen 700 of FIG. 7. The screen depiction 800 provides expenditure data based on vendors within the taxonomy of FIG. 2 by again including the application 230, the big data 240, the calculation 250, the platform 270, the end user 280, the delivery 285, the security and compliance 290, the IT management 295, and the non-IT (not shown) as shown in FIG. 2. Specifically, the size 810 of each vendor box is assigned based on the expenditure level within the taxonomy. Colors such as red and green may be used to indicate increased expenditure (red) and decreased expenditure (green), and the shade of each color may be used to indicate the degree of increase or decrease. Further, resellers who may be involved in the chain may be described by the identified vendor. As is clear in comparison with the depictions of FIGS. 8 and 9, the vendor boxes replace the expenditure values depicted in FIG. 7. Each of the vendors that make up a part of the taxonomy of FIG. 2 is represented by its size and color that provide the meaning inherent in the data, as described herein.

[0081] Figure 10 shows a screen depiction 1000 of the enterprise system 100 of FIG. 1, showing vendor expenditures for a specific category selected in FIG. 7 or FIGS. 8 and 9. In this depiction, the calculation 250, which is part of the taxonomy, is selected and shown. The screen depiction 1000 provides vendor expenditures for the selected calculation 250. As is clear from comparison with FIG. 9, for example, this expands the display of the underlying vendors corresponding to the calculation taxonomy 250 and the calculation 250 category. The screen depiction 1000 emphasizes details about the benefits of vendor integration that can be achieved.

[0082] Figure 11 shows additional details from Figure 10 and the selection of the exemplary calculation 250 taxonomy, and further shows the screen depiction 1100 of the enterprise system 100 of Figure 1 that provides details by selecting a specific vendor 1110 in the depiction of Figure 10. For example, by selecting VMware 1110, additional details regarding the top spenders identified by name among the enterprises in this category and the total expenditure amount in the category are provided. Such information may provide information regarding the responsible persons within the enterprise for a given vendor.

[0083] Figure 12 shows the screen depiction 1200 of the enterprise system 100 of Figure 1 that shows the additional details provided by selecting the vendor 1110 identified in the category 250 of Figure 10. This enables the display of specific orders 1210 in category 250 and vendor 1110. Also, through interaction with the screen 1220, additional information 1230 regarding a specific order, such as order number, date, cost, product, etc., may also be provided to the user.

[0084] Figures 13 and 14 show the screen depiction 1300 of the enterprise system 100 of Figure 1 that shows user-selectable toggles 1305 that enable depicting expenditure trends 1310, strategic value 1320, vendor risk 1330, preferred vendors 1340, and exchange rates 1350 by means of a depiction. The toggles 1305 enable the selection of other metadata and providing it to the user. This visualization system may provide an output depiction of any other metadata found within the system 100. For example, the risk 1330 of the vendors in the taxonomy of Figure 2, which again includes the application 230, big data 240, calculation 250, platform 270, end user 280, delivery 285, security and compliance 290, IT management 295, and non-IT (not shown) as represented in Figure 7, may be displayed and identified by color by the selection of this additional metadata by the toggles 1305.

[0085] FIG. 15 shows a screen depiction 1500 of enterprise system 100 of FIG. 1, showing actual expenditures 1520 of a particular vendor 1510 regardless of category. As depicted, each vendor 1510 is identified and presented based on the amount of expenditure 1520, taking into account the associated organizations inherent as described herein. The amount of expenditure 1520 is represented via the size of the box, and the trend direction of expenditure 1520 is represented by color identification.

[0086] FIG. 16 shows a screen depiction 1600 of enterprise system 100 of FIG. 1, showing the depiction when proceeding to actual expenditures of a particular vendor 1610. This shows the subordinate expenditures that occur for a vendor in an enterprise.

[0087] Although features and elements have been described in a particular combination, it will be understood by those skilled in the art that each feature or element can be used alone or in any combination with other features and elements. Also, the methods described herein may be implemented in a computer program, software, or firmware incorporated into a computer-readable medium for execution by a computer or processor. Examples of computer-readable media include electronic signals (transmitted through wired or wireless connections) and computer-readable storage media. Examples of computer-readable storage media include, but are not limited to, read-only memory (ROM), random access memory (RAM), registers, cache memory, semiconductor memory devices, magnetic media such as internal hard disks and removable disks, magneto-optical media, and optical media such as CD-ROM disks and digital versatile disks (DVDs). A processor associated with software may be used to implement the systems and methods described herein.

Claims

1. An intelligent enterprise system for an enterprise, comprising: an input / output (IO) interface configured to receive information related to accounting information; a processor interconnected with a memory for storing the received information from the IO interface and configured to analyze the received information to provide a timely vendor expenditure analysis; a display device configured to display the provided timely vendor expenditure analysis; wherein: the received information includes the enterprise's expenditure records; the timely vendor expenditure analysis is composed of the enterprise's expenditure records, identifies a plurality of expenditure categories with the direction and magnitude of expenditures associated with each category, and shows the performance of vendors including vendor metadata; in analyzing the received information, the processor: uses the expenditure records to classify the enterprise's vendor expenditures into a plurality of expenditure categories; aggregates the enterprise's vendor expenditures by expenditure category; calculates an expenditure analysis for the vendor based on the value of the vendor's performance criteria. A system characterized by the above.

2. The received information includes category-based expenditures. The system according to claim 1, characterized by the above.

3. The received information includes vendor-based expenditures. The system according to claim 1, characterized by the above.

4. The received information includes expenditures over a period. The system according to claim 1, characterized by the above.

5. The period is a selected fiscal year. The system according to claim 4, characterized by the above.

6. The processor processes the vendor names included in the expenditure records of the received information using nearest neighbor trend scoring to find a vendor in the organizational database that exactly matches the vendor name. The system according to claim 1, characterized by the above.

7. The processor processes the vendor names included in the expenditure records of the received information using a fuzzy matching algorithm to identify the vendor in the organizational database that matches the vendor name. The system according to claim 1, characterized by the above.

8. The processor processes the vendor name included in the expenditure record of the received information using at least two fuzzy matching algorithms to identify the vendor in the organizational database that matches the vendor name. The system according to claim 1, characterized in that.

9. The matching is based on a cutoff score. The system according to claim 7 or claim 8, characterized in that.

10. The system according to claim 1, wherein The system further comprises a database for storing a hierarchical table, The hierarchical table is linked to itself to form a parent-child hierarchy, Using the hierarchical table to identify parent company information for at least one vendor. A system characterized by that.

11. Text tokenization is performed. The system according to claim 1, characterized in that.

12. Based on the account items of the enterprise's expenditure records included in the received information, learn the OEM, In this learning, use the item description of the account item to train a model for determining the OEM. The system according to claim 1, characterized in that.

13. To determine the taxonomy category based on the account items of the enterprise's expenditure records included in the received information, perform automatic classification of the account items into appropriate taxonomy categories. The system according to claim 1, characterized in that.

14. In the automatic classification, use at least one of a hybrid model combining a support vector machine (SVM), a random forest (RF), a neural network (NN), and an expert system (ES) to determine the appropriate taxonomy category from the text description of the account item. The system according to claim 13, characterized in that.

15. The display device is configured to select vendor actual expenditure, peer comparison, market consideration, and regulatory consideration aspects. The system according to claim 1, characterized in that.

16. A method for providing an analysis of a company's vendor expenditures, comprising: Receiving information regarding accounting information via an input / output (IO) interface; Storing the received information via a processor interconnected with a memory, and analyzing the received information to provide a timely vendor expenditure analysis; Displaying the provided timely vendor expenditure analysis via a display device. A method characterized by: The received information includes the expenditure records of the enterprise, The timely vendor expenditure analysis is composed of the expenditure records of the enterprise, identifies a plurality of expenditure categories in which the direction and magnitude of expenditures for each category are associated, and shows the performance of the vendors including vendor metadata, In the analysis of the received information, Using the expenditure records, classify the vendor expenditures of the enterprise into a plurality of expenditure categories, Aggregate the vendor expenditures of the enterprise by expenditure category, Calculate the expenditure analysis for the vendor based on the value of the performance criteria of the vendor. A method characterized by the above. **Claim 17** The received information includes at least one of category-based expenditures and vendor-based expenditures. The method according to claim 16, characterized by the above. **Claim 18** The analysis of the received information includes Processing by the processor to find a vendor in the organizational database that exactly matches the vendor name included in the expenditure records of the received information using nearest neighbor trend scoring, and At least one of the processing by the processor to identify the vendor in the organizational database that matches the vendor name using a fuzzy matching algorithm. The method according to claim 16, characterized by the above. **Claim 19** Train the OEM based on the ledger items of the expenditure records of the enterprise included in the received information. In this training, use the item descriptions of the ledger items to train a model for determining the OEM. Automatically classify the appropriate taxonomic categories of the ledger items in order to determine the taxonomic categories based on the ledger items of the expenditure records of the enterprise included in the received information. The method according to claim 16, characterized by the above. **Claim 20** In the analysis, use at least one of a hybrid model combining a support vector machine (SVM), a random forest (RF), a neural network (NN), and an expert system (ES) to determine the taxonomic category of the ledger item from the text description of the ledger item of the expenditure records of the enterprise. The method according to claim 16, characterized by the above.

Images