# US 2019/0244134 A1 · Intelligent service request classification and assignment

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0244134 A1
- Document ID: US20190244134A1
- Status: Published application
- Kind: A1
- Application number: 16/386,550
- Filed: 2019-04-17
- Published: 2019-08-08
- Earliest claimed: 2015-05-29
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Tyson R. Midboe
- Web page: https://patentgrep.com/patent/US20190244134A1

## CPC

- G06N 20/00: Machine learning
- H04L 67/535: {Tracking the activity of the user (network monitoring arrangements H04L43/00; recording of computer activity G06F11/34)}
- H04L 67/63: Routing a service request depending on the request content or context
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- H04L 41/142: using statistical or mathematical methods
- H04L 41/5045: {Making service definitions prior to deployment}
- H04L 67/60: Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources (admission control or resource allocation H04L47/70)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Approaches presented herein enable intelligent service request classification and assignment learning. More specifically, a request comprising a free form text or spoken description is received from a user. The request description is parsed and classified by a regression-based classifier. The regression-based classifier classifies based on, for example: the description itself; the requestor's history of requests, and/or supplemental demographics about a requestor. Optionally, a user may verify the classification or select from a plurality of returned classifications. A service provider or administrator confirms that a classification is correct. If not, the incorrectly classified request is queued. If so, the correctly classified request is added to a set of training data to be used in classifying future requests.

## Claims

### Claim 1

A method of intelligent service request classification and assignment learning, comprising:
receiving, at an automated user assistance system, a single description of a service request that is freeform in a natural language from a user;
converting the description and an identification of the user into a set of feature vectors;
applying to the service request description a logistic regression algorithm configured to determine a classification of the service request based on the description and a historic frequency of classified service requests submitted by the user in the set of feature vectors;
forwarding the service request to a human administrator based on the classification;
receiving a determination from the human administrator whether the service request is correctly classified; and
continuously training the logistic regression algorithm against an increasingly refined set of training data of correctly classified service requests.

### Claim 2 (depends on 1)

The method of claim 1, wherein the logistic regression algorithm is a naïve Bayes classifier.

### Claim 3 (depends on 1)

The method of claim 1, where, in the case that the service request is correctly classified, the method further comprises storing an identification of the user and a classification of the service request in the set of training data.

### Claim 4 (depends on 3)

The method of claim 3, further comprising:
calculating a frequency the user requests a service of the classification; and
storing the frequency in the set of training data.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
predicting a probability a classification determined with the logistic regression algorithm applies to the service request; and
in the case that the probability is within a threshold, assigning the classification to the service request automatically.

### Claim 6 (depends on 5)

The method of claim 5, where, in the case that the probability is under a threshold, the method further comprises:
returning a set of one or more classification options to the user; and
assigning a classification option selected by the user to the service request.

### Claim 7 (depends on 1)

The method of claim 1, wherein the logistic regression algorithm is further configured to determine a classification of the service request based on supplemental demographics about the user, the supplemental demographics being collected from one or more of: a human resources (HR) system, a configuration management database (CMDB), a Customer Relationship Management (CRM) system, an Enterprise Resource Planning (ERP) system, and a social media system.

### Claim 8

A computer system for intelligent service request classification and assignment learning, the computer system comprising:
a memory device comprising program instructions;
a bus coupled to the memory device; and
a processor, for executing the program instructions, coupled to a request management tool via the bus that when executing the program instructions causes the system to:
receive, at an automated user assistance system, a single description of a service request that is freeform in a natural language from a user;
converting the description and an identification of the user into a set of feature vectors;
apply to the service request description a logistic regression algorithm configured to determine a classification of the service request based on the description and a historic frequency of classified service requests submitted by the user in the set of feature vectors;
forward the service request to a human administrator based on the classification;
receive a determination from the human administrator whether the service request is correctly classified; and
continuously train the logistic regression algorithm against an increasingly refined set of training data of correctly classified service requests.

### Claim 9 (depends on 8)

The system of claim 8, wherein the logistic regression algorithm is a naïve Bayes classifier.

### Claim 10 (depends on 8)

The system of claim 8, where, in the case that the service request is correctly classified, the program instructions further cause the system to store an identification of the user and a classification of the service request in the set of training data.

### Claim 11 (depends on 10)

The system of claim 10, the program instructions further causing the system to:
calculate a frequency the user requests a service of the classification; and
store the frequency in the set of training data.

### Claim 12 (depends on 8)

The system of claim 8, the program instructions further causing the system to:
predict a probability a classification determined with the logistic regression algorithm applies to the service request; and
in the case that the probability is within a threshold, assign the classification to the service request automatically.

### Claim 13 (depends on 12)

The system of claim 12, where, in the case that the probability is under a threshold, the program instructions further cause the system to:
return a set of one or more classification options to the user; and
assign a classification option selected by the user to the service request.

### Claim 14 (depends on 8)

The system of claim 8, wherein the logistic regression algorithm is further configured to determine a classification of the service request based on supplemental demographics about the user, the supplemental demographics being collected from one or more of: a human resources (HR) system, a configuration management database (CMDB), a Customer Relationship Management (CRM) system, an Enterprise Resource Planning (ERP) system, and a social media system.

### Claim 15

A computer program product for intelligent service request classification and assignment learning, the computer program product comprising a computer readable storage device, and program instructions stored on the computer readable storage device, to
receive, at an automated user assistance system, a single description of a service request that is freeform in a natural language from a user;
convert the description and an identification of the user into a set of feature vectors;
apply to the service request description a logistic regression algorithm configured to determine a classification of the service request based on the description, a historic frequency of classified service requests submitted by the user, and supplemental demographics about the user in the set of feature vectors;
forward the service request to a human administrator based on the classification;
receive a determination from the human administrator whether the service request is correctly classified; and
continuously train the logistic regression algorithm against an increasingly refined set of training data of correctly classified service requests.

### Claim 16 (depends on 15)

The computer program product of claim 15, wherein the logistic regression algorithm is a naïve Bayes classifier.

### Claim 17 (depends on 15)

The computer program product of claim 15, where, in the case that the service request is correctly classified, the program instructions are further configured to store an identification of the user and a classification of the service request in the set of training data.

### Claim 18 (depends on 17)

The computer program product of claim 17, the program instructions being further configured to:
calculate a frequency the user requests a service of the classification; and
store the frequency in the set of training data.

### Claim 19 (depends on 15)

The computer program product of claim 15, the program instructions being further configured to:
predict a probability a classification determined with the logistic regression algorithm applies to the service request; and
in the case that the probability is within a threshold, assign the classification to the service request automatically.

### Claim 20 (depends on 19)

The computer program product of claim 19, where, in the case that the probability is under a threshold, the program instructions are further configured to:
return a set of one or more classification options to the user; and
assign a classification option selected by the user to the service request.