Method for making cognitive bidding decision

Patent metadata

Field	Value
Publication number	US-11113756-B2
Application number	US-201816140084-A
Country	US
Kind code	B2
Filing date	Sep 24, 2018
Priority date	Sep 24, 2018
Publication date	Sep 7, 2021
Grant date	Sep 7, 2021

Abstract

A method for a provider to make a cognitive bidding decision may include using a processor and associated memory to a) obtain information, including unstructured data, relating to deals with procuring organizations, and b) determine a respective predicted business growth of each procuring organization based upon the information. The method may also include using the processor and the associated memory to c) determine a respective predicted growth of provider services for each procuring organization based upon the information, and d) determine a respective probability of winning each bid for each deal from each procuring organization. The method may further include using the processor and the associated memory to e) determine a respective predicted revenue based upon winning each bid, and determine whether to submit each bid based upon b), c), d), and e).

First claim

What is claimed is:

1. A method for a provider to make a cognitive bidding decision comprising:
using a processor and associated memory to
obtain information, including unstructured data and structured data, relating to a plurality of deals with a plurality of procuring organizations;
train a first machine learning prediction model based upon the structured data;
train a second machine learning prediction model based upon the unstructured data;
train a combined machine learning model, the combined machine learning model configured to provide a final output based upon at least one of an aggregation of an output of each of the first and second machine learning prediction models and a building of a model based upon the output of each of the first and second machine learning prediction models;
determine a respective predicted business growth of each procuring organization based upon the information;
determine a respective predicted growth of provider services for each procuring organization based upon the information;
determine a respective probability of winning each bid for each deal from each procuring organization;
determine a respective predicted revenue based upon winning each bid; and
determine whether to submit each bid based upon the respective predicted business growth of each procuring organization, the respective predicted growth of provider services for each procuring organization, the respective probability of winning each bid for each deal from each procuring organization, and the respective predicted revenue;
wherein one of the respective predicted business growth and the respective predicted growth of provider services is determined based upon the first machine learning prediction model based upon the structured data, the second machine learning prediction model based upon the unstructured data, and the combined machine learning model; and
wherein each respective predicted business growth is one of multiple levels of growth and wherein the first machine learning prediction model generates a prediction percentage for each of the multiple levels of growth.

2. The method of claim 1 comprising using the processor and associated memory to further determine whether to submit each bid based upon available provider resources.

3. The method of claim 1 wherein using the processor and associated memory to determine whether to submit each bid is based upon a highest predicted revenue.

4. The method of claim 1 wherein using the processor and associated memory to determine the respective predicted revenue is based upon the respective predicted business growth of each procuring organization, the respective predicted growth of provider services for each procuring organization, the respective probability of winning each bid for each deal from each procuring organization, and a respective deal revenue for each deal.

5. The method of claim 1 wherein using the processor and associated memory to determine the respective predicted business growth of each procuring organization comprises using feature vector build up and model training.

6. The method of claim 1 wherein using the processor and associated memory to determine the respective predicted growth of provider services for each procuring organization comprises using feature vector build up and model training.

7. The method of claim 1 comprising using the processor and associated memory to determine the respective probability of winning each bid comprises using a machine learning classification model.

8. The method of claim 1 wherein the unstructured data comprises at least one of news data, expert rating data, growth expectation data, provider-client communications, requests for proposal, and procuring organization website information.

9. The method of claim 1 wherein the information comprises historical deal data and procuring organization metadata.

10. The method of claim 1 wherein one of the respective predicted business growth and the respective predicted growth of provider services is determined based upon structuring of the unstructured data to generate newly structured data and generation of a trained prediction model based upon the structured data and the newly structured data.

11. A system for a provider to make a cognitive bidding decision comprising:
a processor and associated memory configured to
obtain information, including unstructured data, relating to a plurality of deals with a plurality of procuring organizations;
train a first machine learning prediction model based upon the structured data;
train a second machine learning prediction model based upon the unstructured data;
train a combined machine learning model, the combined machine learning model configured to provide a final output based upon at least one of an aggregation of an output of each of the first and second machine learning prediction models and a building of a model based upon the output of each of the first and second machine learning prediction models;
determine a respective predicted business growth of each procuring organization based upon the information;
determine a respective predicted growth of provider services for each procuring organization based upon the information;
determine a respective probability of winning each bid for each deal from each procuring organization;
determine a respective predicted revenue based upon winning each bid; and
determine whether to submit each bid based upon the respective predicted business growth of each procuring organization, the respective predicted growth of provider services for each procuring organization, the respective probability of winning each bid for each deal from each procuring organization, and the respective predicted revenue;
wherein one of the respective predicted business growth and the respective predicted growth of provider services is determined based upon the first machine learning prediction model based upon the structured data, the second machine learning prediction model based upon the unstructured data, and the combined machine learning model; and
wherein each respective predicted business growth is one of multiple levels of growth and wherein the first machine learning prediction model generates a prediction percentage for each of the multiple levels of growth.

12. The system of claim 11 wherein the processor and associated memory are configured to determine whether to submit each bid based upon available provider resources.

13. The system of claim 11 wherein the processor and associated memory are configured to determine whether to submit each bid based upon a highest predicted revenue.

14. The system of claim 11 wherein the processor and associated memory are configured to determine the respective predicted revenue based upon the respective predicted business growth of each procuring organization, the respective predicted growth of provider services for each procuring organization, the respective probability of winning each bid for each deal from each procuring organization and a respective deal revenue for each deal.

15. The system of claim 11 wherein the processor and associated memory are configured to determine the respective predicted business growth of each procuring organization using feature vector build up and model training.

16. A non-transitory computer readable medium for a provider to make a cognitive bidding decision, the computer readable medium comprising computer executable instructions which when executed by a processor cause the processor to perform operations comprising:
obtaining information, including unstructured data, relating to a plurali

Assignees

• IBM

Inventors

• Megahed Aly
• Mohamed Mohamed
• Tata Samir

Classifications

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06Q30/08 Primary

  Auctions · CPC title

• G06N20/10

  using kernel methods, e.g. support vector machines [SVM] · CPC title

• G06N20/00

  Machine learning · CPC title

• G06Q30/0201 Primary

  Market modelling; Market analysis; Collecting market data · CPC title