Abstract: A machine learning based (ML-based) method and system for automatically extracting and correcting financial information from documents, is disclosed. Initially, the documents are obtained from data sources and pre-processed to generate the pre-processed data associated with contents within the document. The contents are classified as potential key-value pairs corresponding to the financial information based on the system prompts and extracted using the ML model. The potential key-value pairs are corrected to obtain the corrected key-value pairs based on custom prompts, using the ML model. The corrected key-value pairs corresponding to the financial information are provided as the output to the end users on user interfaces associated with an electronic device. This technique extracts financial information regardless of structure or alignment by learning to recognize any added or removed prefixes or suffixes, enabling the prefixes or suffixes to make corrections and generate accurate key-value pairs.

Abstract: A system and method for predicting financial transaction patterns is disclosed. The method includes receiving invoice data of one or more customers. The method further includes receiving granularity levels, thereby generating a set granularity level instances based on various invoice data attributes. The method further computes payment frequency bucket features for all the generated set of granularity level instances and assigns invoices to clusters based on the set of payment frequency bucket features. The set of payment frequency bucket features assigns a customer to be one of: a weekly payer, alternative weekly payer, a monthly payer, a bi-monthly, a quarterly payer, a half yearly payer and an annual payer. The method further includes the generation of a set of payment pattern features based on the set of payment frequency bucket features of the cluster. Further, the method includes selection of an optimal pattern with highest probability of adherence.

Abstract: A machine learning based computing method for generating electronic mails for accounts receivables management is disclosed.

Abstract: Disclosed embodiments provide tools and techniques for the automated generation of a series of sets of collection worklists for collection agents. Consecutive sets of collection worksheets may be created in a series over any selected time interval, for example weekly, daily, or monthly. Worklists for one time interval can be combined into worklists for other time intervals. The generation of a collection worklist for a selected time interval can include several processing and monitoring steps implemented with machine learning or artificial intelligence. The collection worklists may then be provided to collections agents, collections managers, or downstream software modules to inform and guide the collection activities undertaken by collections agents.

Abstract: The present invention is related to data processing methods and systems thereof. According to an embodiment, the present invention provides a method of processing claim deduction documents using a machine learning model. The process begins by accessing data files and extracting information from them, which is subsequently stored. This document information, along with the machine learning model trained on various document formats, is used to classify the data files and generate tabular data. From this tabular data, data objects are created and included in an output data file. The information from the output file is then used to update the data of the machine learning model, optimizing it for improved future document processing. There are other embodiments as well.

Abstract: A Machine Learning (ML)-based computing system and method for financial transaction based customer worklist generation is disclosed. A data determination module configured to obtain a credit sale amount, an account receivable as of a run date of the module (RD), a disputed invoice amount and a skipped invoice amount using an Machine Learning (ML) model. A DSO component calculation module configured to calculate the obtained DSO components for each entity corresponding to a grouping category at a given point of time period. A DSO impact score generation module configured to generate a DSO impact score based on the estimated open amount reduction, desired number of days in period and the credit sale amount. A Machine Learning insight module configured to calculate the generated DSO impact score based on historical customer information associated with one or more customers. A data output module configured to output the DSO impact score.

Abstract: A computer-implemented method for generating interactive actionable electronic mails, is disclosed.

Abstract: A machine learning based computing method for computing optimal contact times for contacting first users including debtors, is disclosed. The machine learning based computing method includes: receiving inputs from second users including debt collectors; extracting data associated with the first users and the second users from databases, based on the inputs received from the second users; computing contact feature scores based on the extracted data associated with the first users and the second users, for each specified interval of a contact prediction window; computing first user call scores for each specified interval of the contact prediction window based on the contact feature scores for each specified interval of the contact prediction window, using a machine learning model; and computing the optimal contact times and a prioritized list of the optimal contact times by ranking each specified interval of the contact prediction window associated with the first user call scores.

Abstract: Novel tools and techniques are provided for detecting and correcting anomalous events in real-time in as users enter data into a general ledger. In various embodiments, a computing system may receive one or more user inputs. The computing system might analyze the one or more user inputs to determine whether there are one or more anomalous transactions associated with the one or more user inputs by obtaining historical information associated with the one or more user inputs and correlating, using one or more machine learning algorithms, the historical information and the one or more user inputs to detect one or more anomalies. Based on the correlation between the historical information and the one or more first user inputs, the computing system might detect at least one anomalous transaction and might generate one or more recommended actions to correct the at least one anomalous transaction.

Abstract: A machine learning based computing method for computing optimal time for sending email, is disclosed. The machine learning based computing method includes steps of: receiving inputs from second users; extracting data associated with the first users and the second users from databases, based on the inputs received from the second users; computing electronic mail feature scores based on the extracted data associated with the first users and the second users, for each specified interval of contact prediction window; computing first user electronic mail scores for each specified interval of the contact prediction window based on the electronic mail feature scores for each specified interval of the contact prediction window, using a machine learning model; and computing at least one of: the optimal times and a prioritized list of the optimal times by ranking each specified interval of the contact prediction window associated with the first user electronic mail scores.

Abstract: A system and method for correcting image data is disclosed. The method includes receiving one or more documents from one or more electronic mediums. The method further includes determining a primary character and one or more alternate characters corresponding to the mis-captured character image, extracting one or more confident instances of the primary character and the one or more alternate characters from the one or more documents and generating one or more scores corresponding to the primary character and the one or more alternate characters. Further, the method includes predicting a correct character corresponding to the mis-captured character image by using a trained image prediction-based ML model and automatically replacing the mis-captured character image with the predicted correct character.

Abstract: A system and method for or facilitating correction of data in documents is disclosed. The method includes receiving one or more documents from and scanning the received one or more documents by using a document processing system for obtaining one or more mis-captured data fields. The method further includes obtaining a historical correction data and determining one or more deltas based on the obtained historical correction data by using a trained data correction-based ML model. Further, the method includes parsing the determined one or more deltas into one or more datasets, generating one or more correct data fields corresponding to the one or more mis-captured data fields and automatically replacing the one or more mis-captured data fields with the generated one or more correct data fields based on one or more predefined rules.

Abstract: An enhanced customer relationship management (CRM) system is provided. The enhanced CRM system performs activities automatically and with the assistance of artificial intelligence and machine learning based on historical information. The enhanced CRM system provides: a) scheduling assistance prior to a customer contact, b) assistance during a call to direct call focus and achieve a personal connection between a customer facing user of the CRM system and a target customer contact, and c) automated assistance to complete a contact and transition to a next target customer contact. Achievement goals for a customer facing user may be presented and monitored with respect to a goal achievement period. Schedules may be dynamically adjusted across multiple CRM system users and with respect to overall organizational goals to enhance achievement of goals.

Abstract: Novel tools and techniques are provided for implementing identification and removal of noise from documents, and, more particularly, to methods, systems, and apparatuses for implementing identification and removal of noise from financial documents using one or more machine learning algorithms. In various embodiments, computing system might receive a document. The computing system might detect, using one or more machine learning algorithms, that noise exists in the document. Based on the detection that noise exists in the document, the computing system might remove the noise from the document. Once the noise is removed from the document, the computing system might generate a copy of the document with the noise removed while retaining important or useful information contained in the document.

Abstract: A system to process virtual credit card (VCC) payments automatically is provided. Electronic delivery of VCC payment information may be transferred as electronic data representing a VCC payment to the seller (e.g., an email). Data exchange methods for VCC payments include application programming interfaces (APIs), email, files, etc. Secure processing is provided via an environment that meets rigorous security standards (e.g., Payment Card Industry (PCI) data security standards). Credentials of the sending entity establish the context that allows the system to associate the data received with the correct buyer and seller. Authentication may be provided within the context of the data transferred. Authentication may also use the sender's email address to validate the sender using a digital signature of the email contents. The system may combine digital signature validation of an email with a validation of a sending entity using the DomainKeys Identified Mail (DKIM) email authentication protocol.

Abstract: Provided techniques manage and predict future events. For example, in a payment implementation, a supplier, at any given point in time, has multiple customer debtors that may owe payments (e.g., have outstanding invoices). Utilizing historical attributes for a given customer debtor payment predictions may be determined. By analyzing outstanding debts associated with this debtor customer an amount owed may be calculated and a predicted payment (e.g., a payment that has not yet been indicated by that debtor customer) created. Events may be provided to a second system to correlate predictions across multiple debtor collectors. Correlated information may be used to predict cash flow needs of an organization. Alternatively, optimization of help desk systems may be provided based on predictions from analysis of multiple events in an Event-driven feed back system. Provided techniques may be generalized to other applications as well.

Abstract: A system to process virtual credit card (VCC) payments automatically is provided. Electronic delivery of VCC payment information may be transferred as electronic data representing a VCC payment to the seller (e.g., an email). Data exchange methods for VCC payments include application programming interfaces (APIs), email, files, etc. Secure processing is provided via an environment that meets rigorous security standards (e.g., Payment Card Industry (PCI) data security standards). Credentials of the sending entity establish the context that allows the system to associate the data received with the correct buyer and seller. Authentication may be provided within the context of the data transferred. Authentication may also use the sender's email address to validate the sender using a digital signature of the email contents. The system may combine digital signature validation of an email with a validation of a sending entity using the DomainKeys Identified Mail (DKIM) email authentication protocol.

Abstract: A system generates trade deduction settlement rules and associated confidence scores independent of buyer specifications. A machine learning equipped rewards based method performed by the system analyzes historically matched deductions and promotions to understand patterns. Penalties are applied to outdated rules, and recent trends are promoted through rewards. All available deduction-promotion combinations may be analyzed in batches for a given time period at each pair level within an artificial intelligence model of the method. A rules selector selects the most recurring patterns along those combinations based upon definable thresholds. The system finds hidden patterns to provide suggestions for trade deduction settlement. The system further captures the rules and evolves the rules over time.

Abstract: An enhanced customer relationship management (CRM) system is provided. The enhanced CRM system performs activities automatically and with the assistance of artificial intelligence and machine learning based on historical information. The enhanced CRM system provides: a) scheduling assistance prior to a customer contact, b) assistance during a call to direct call focus and achieve a personal connection between a customer facing user of the CRM system and a target customer contact, and c) automated assistance to complete a contact and transition to a next target customer contact. Achievement goals for a customer facing user may be presented and monitored with respect to a goal achievement period. Schedules may be dynamically adjusted across multiple CRM system users and with respect to overall organizational goals to enhance achievement of goals.