Method and system for prioritizing communications based on sentence classifications

• US 8,112,268 B2
• Filed: 10/20/2008
• Issued: 02/07/2012
• Est. Priority Date: 08/31/2004
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Chen, Zheng, Ma, Wei-Ying, Zeng, Hua-Jun, Zhang, Benyu
• Primary Examiner(s): Han, Qi
• Application Number:
  US12/254,796
• Publication Number:
  US 20090106019A1
• Time in Patent Office: 1,205 Days
• Field of Search: 704/4, 704/1, 704/10, 704/9
• US Class Current: 704/4
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 16/353 into predefined classes

Abstract

A method and system for prioritizing communications based on classifications of sentences within the communications is provided. A sentence classification system may classify sentences of communications according to various classifications such as “sentence mode.” The sentence classification system trains a sentence classifier using training data and then classifies sentences using the trained sentence classifier. After the sentences of a communication are classified, a document ranking system may generate a rank for the communication based on the classifications of the sentences within the communication. The document ranking system trains a document rank classifier using training data and then calculates the rank of communications using the trained document rank classifier.

Claims

1. A computer-readable medium containing instructions for controlling a computer system to rank documents, by a method comprising:

providing a sentence classifier to classify sentences into classifications of sentences;

training a document rank classifier by representing each document of training data by the classifications of its sentences as determined by the sentence classifier and a rank of each document;

representing a document by the classifications of its sentences as determined by the sentence classifier; and

applying the trained document rank classifier to the representation of the document to determine the rank of the document.

2. The computer-readable medium of claim 1 wherein the documents are electronic mail messages and including prioritizing electronic mail messages based on their ranks.

3. The computer-readable medium of claim 1 wherein the sentences are classified based on sentence mode.

4. The computer-readable medium of claim 1 wherein the sentence classifications include declarative, imperative, interrogative, and exclamatory.

5. The computer-readable medium of claim 1 wherein the sentence classifications include appreciative and attach.

6. The computer-readable medium of claim 1 wherein the training includes applying an Ada-boosting algorithm.

7. The computer-readable medium of claim 1 wherein the training includes applying a support vector machine algorithm.

8. The computer-readable medium of claim 1 wherein the providing of the sentence classifier includes:

providing training sentences and their classifications;

generating feature representations of the sentences;

training a sentence classifier to classify sentences based on the generated feature representations and the classifications of the sentences they represent;

receiving a sentence;

generating a feature representation of the sentence; and

applying the trained sentence classifier to the generated feature representation of the sentence to classify the sentence.

9. A computer system for ranking documents, comprising:

a memory storing computer-executable instructions of;

a sentence classifier;

a document rank classifier trained based on classifications of sentences of documents, the classifications being generated using the sentence classifier, and trained based on ranks of documents; and

a component that classifies sentences of a document using the sentence classifier, represents the document by classifications of its sentences as determined by the sentence classifier, and applies the document rank classifier to the classified sentences to generate a rank for the document; and

a processor that executes the computer-executable instructions stored in the memory.

10. The computer system of claim 9 wherein the documents are electronic mail messages.

11. The computer system of claim 9 wherein the sentences are classified based on sentence mode.

12. The computer system of claim 9 wherein the sentence classifications include declarative, imperative, interrogative, and exclamatory.

13. The computer system of claim 12 wherein the sentence classifications include appreciative and attach.

14. The computer system of claim 9 wherein the classifiers are trained using an Ada-boosting algorithm.

15. The computer system of claim 9 wherein the classifiers are trained using a support vector machine algorithm.