Automatic organization of documents through email clustering

• US 7,765,212 B2
• Filed: 12/29/2005
• Issued: 07/27/2010
• Est. Priority Date: 12/29/2005
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Renshaw, Erin L., Surendran, Arungunram C., Platt, John C.
• Primary Examiner(s): Trujillo; James
• Assistant Examiner(s): Casanova; Jorge A
• Application Number:
  US11/321,963
• Publication Number:
  US 20070156732A1
• Time in Patent Office: 1,671 Days
• Field of Search: 707/101, 707/3, 707/737, 707/738, 707/739, 707/999.003, 707/999.101, 704/245
• US Class Current: 707/738
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: H04L 51/00 User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail

Abstract

A system that facilitates organization of emails comprises a clustering component that clusters a plurality of emails and creates topics for emails by assigning key phrases extracted from emails within one or more clusters. An organization component then utilizes the key phrases to organize documents. Furthermore, the organization component can comprise a probability component that determines a probability that a document belongs to a certain topic.

Claims

1. A system comprising:

computer-readable storage media having processor-executable instructions embodied therein; and

a processor operatively coupled to the computer-readable storage media to execute the processor-executable instructions for implementing computer-executable components comprising;

a clustering component that receives and clusters a plurality of emails, the clustering component automatically determining and creating topics for the emails by assigning key phrases extracted from the emails within one or more clusters, the topics being cohesive concepts relevant to a user associated with the emails;

an organization component that utilizes the topics created from the emails to organize the emails and to also organize other documents of the user within a graphical user interface, the other documents having been stored in a data store separately from the emails and comprising at least one of word processing documents, spreadsheets, presentation files, video files, audio files or digital images; and

an interface component that displays the topics defined by the clustering component and information pertaining to the organized emails and the other documents associated with the topics.

2. The system of claim 1, the organization component organizing the other documents by comparing content of each other document with the key phrases extracted from the one or more clusters for associating each other document with one or more of the topics.

3. The system of claim 1, further comprising a filtering component that removes a cluster defined by the clustering component if a number of emails within the cluster is below a predefined threshold.

4. The system of claim 1, the organization component comprises a probability component that determines a probability that a document belongs to a certain topic and assigns the document to the certain topic.

5. The system of claim 4, the clustering component employs Probabilistic Latent Semantic Analysis to automatically characterize each topic by a distribution of words within clusters.

6. The system of claim 4, the probability component determines a probability that a word belongs to a certain topic.

7. The system of claim 1, further comprising a key phrase extraction component that characterizes the topic by selecting descriptive keywords from a cluster and labeling the cluster with such descriptive keywords.

8. The system of claim 1,

wherein the plurality of emails comprise a first plurality of emails stored in a first storage device in association with a web-based email account, and a second plurality of emails stored in a second storage device associated with an email application installed on a computer including the processor,

wherein the clustering component receives both the first plurality of emails and the second plurality of emails for clustering.

9. The system of claim 1, further comprising:

an analysis component that reviews text in subject lines of the emails and creates a list of noun phrases;

a supplement component that supplements the noun phrases with a set of words that appear in the noun phrases; and

a counting component that counts a number of occurrences of the supplemented noun phrases in a body of an email.

10. The system of claim 1, further comprising a token component that supplies tokens to each email within the plurality of emails corresponding to deliverers of the emails and recipients of the emails, the tokens employed by the clustering component to refine a topic and to separate topics that are lexically similar.

11. The system of claim 1, the interface component automatically arranges the topics defined by the clustering component and presents such arrangement to a user.

12. The system of claim 11, the interface component comprises a relevance calculator component that calculates a relevance measure of an email with respect to a topic that includes the email, documents within a topic can be arranged within the graphical user interface based at least in part upon the calculated relevance measure.

13. The system of claim 1, further comprising an assignment component that assigns an incoming email to at least one topic based upon a calculated probability of relevance to the topic.

14. A method comprising:

employing a processor executing computer executable instructions stored on a computer-readable storage medium to implement the following acts;

receiving a plurality of emails;

clustering the plurality of emails into multiple clusters;

performing key phrase extraction upon emails within at least one of the clusters;

characterizing a topic with one or more extracted key phrases, the topic being a cohesive concept that is relevant to a user associated with the plurality of emails, the topic being at least one of;

an activity in which the user participates, an event the user organized or attended, a person or group of people within an organization to which the user belongs, or a project; and

automatically organizing non-email documents of the user stored in a first data store and the plurality of emails based upon the topics characterized with the one or more extracted key phrases from the emails, the non-email documents stored in the data store being organized by comparing content of each non-email document with the key phrases extracted from the multiple clusters of the plurality of emails for associating each non-email document with one or more of the topics, the non-email documents comprising at least one of;

word processing documents, spreadsheets, presentation files, video files, audio files or digital images.

15. The method of claim 14, further comprising supplying tokens to each email within the plurality of emails corresponding to deliverers of the emails and recipients of the emails, the tokens employed to refine a topic and to separate topics that are lexically similar.

16. The method of claim 15, the act of clustering further comprises:

representing the plurality of emails using term frequency-inverse document frequency (tf-idf) vectors of selected words;

employing a cosine distance measure to measure email similarity;

creating cluster initializations by performing agglomerative clustering on a subset of the plurality of emails based on the measure of email similarity; and

running K-means using the cluster initializations on each of the plurality of emails.

17. The method of claim 14,

wherein receiving the plurality of emails comprises receiving a first plurality of emails stored in a first data store, wherein the first plurality of emails are associated with an email application installed on a computer including the processor,

wherein receiving the plurality of emails further comprises receiving a second plurality of emails stored in a second data store in association with a web-based email account,

wherein both the first plurality of emails and the second plurality of emails are clustered during the clustering.

18. The method of claim 14, further comprising:

displaying a list of expandable topics upon a graphical user interface, the topics correlate to labeled clusters; and

displaying emails associated with the expandable topics upon selection of at least one of the expandable topics.

19. A computer-implemented method comprising:

employing at least one processor that executes computer executable code stored in computer-readable storage media to effect the following;

receiving a first plurality of emails stored in a first storage device in association with a web-based email account;

receiving a second plurality of emails stored in a second storage device associated with an email application installed on a computer including the processor;

clustering the first and second plurality of emails, the clustering employing multistage clustering that runs a first cluster technique on the first and second plurality of emails to form one or more clusters, and then runs a second cluster technique on the one or more clusters initialized from the first technique, the second cluster technique being different from the first cluster technique and further refining the one or more clusters so as to facilitate key phrase extraction from emails within the one or more clusters;

extracting key phrases from the emails within the one or more clusters and labeling each of the one or more clusters with a subset of the extracted key phrases for establishing a topic for each of the one or more clusters;

performing post processing on the topics to remove certain key phrases associated with the topics based on a determination that the certain key phrases occur in multiple emails and are not representative of a topic;

organizing non-email documents and the emails within a graphical user interface based upon the topics generated from the clustered emails, the non-email documents being organized by comparing content of each non-email document with the key phrases extracted from the multiple clusters of the plurality of emails for associating each non-email document with one or more of the topics, the non-email documents comprising at least one of;

word processing documents, spreadsheets, presentation files, video files, audio files or digital images; and

rendering information related to the organized non-email documents and emails.

20. The method according to claim 19 further comprising:

running K-means clustering as the first cluster technique to create one or more clusters; and

running latent semantic analysis as the second cluster technique that further refines the one or more clusters to facilitate key phrase extraction from emails within the one or more clusters.