Data ingestion module for event detection and increased situational awareness

• US 9,984,427 B2
• Filed: 12/02/2014
• Issued: 05/29/2018
• Est. Priority Date: 12/02/2013
• Status: Active Grant

• Current Assignee: Finch Computing LLC (Qbase, LLC)
• Original Assignee: Qbase, LLC
• Inventors: Dave, Rakesh, Boddhu, Sanjay
• Primary Examiner(s): Singh, Amresh
• Application Number:
  US14/557,781
• Publication Number:
  US 20150154249A1
• Time in Patent Office: 1,274 Days
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: G06Q 50/10 Services

Abstract

A system and method for detecting and summarizing events based on data feeds from a plurality of sources. Such sources may include social media networks, text messages, news feeds among others. The system may receive raw information from such sources containing data related with possible events. Method for event detection may include pre-processing and normalizing data input from any source registered, this may also include; extracting and disambiguating events and entities, associate event and entities, correlate events and entities associated from a data input which results from a different data source, for validating/verifying an event. Subsequently, the validated/verified event may be stored in a local data storage and/or in a web-server.

Claims

1. A computer-implemented method comprising:

receiving, by a computer, from a plurality of data sources, a plurality of raw data streams containing one or more events, wherein each respective data source comprises a server publishing a corresponding raw data stream;

pre-processing, by the computer, the plurality of raw data streams for subsequent processing, wherein pre-processing the plurality of raw data streams comprises;

responsive to identifying, by the computer, one or more ingestion modules corresponding respectively to a set of one or more raw data streams of the plurality of raw data streams;

determining, by the computer, whether each respective raw data stream in the set satisfies a set of one or more specific criteria based on comparing each respective raw data stream in the set against one or more data files containing data of one or more types stored in a database, wherein at least one criterion of the one or more specific criteria is selected from the group consisting of;

type of vocabulary used, data format, limitation of characters, and frequency of new data streams released, and wherein at least one data file of the one or more data files is an event template modeling an event; and

responsive to identifying, by the computer, a second set of one or more raw data streams not corresponding to the one or more ingestion modules;

determining, by the computer, whether each respective raw data stream in the second set satisfies a set one or more common criteria based on comparing each respective raw data stream in the second set against one or more data files containing data of one or more types stored in the database, wherein the comparison is based on at least one type of data selected from the group consisting of;

templates, models, search paths, and definitions, and wherein at least one data file of the one or more data files is a second event template modeling a second event;

processing, by the computer, each respective raw data stream of the second set as indicating a new event responsive to determining that the respective raw data stream satisfies the common set of criteria,

determining, by the computer, whether one or more features in each respective data stream indicated as the new event satisfy one or more event models in a categorization table, based upon the computer comparing the one or more features of each respective data stream as the new event against the one or more event models; and

responsive to the computer determining that the one or more features from each respective data stream indicated as the new event fail to satisfy at least one event model in at least one categorization table;

comparing, by the computer, the one or more features from each respective data stream indicated as the new event against one or more uncategorized event models in an uncategorized event table, wherein the uncategorized event table store records associated with new unknown event models; and

storing, by the computer, the one or more features from each respective data stream indicated as the new event as a new uncategorized event model in the uncategorized event table, in response to determining the one or more features from each respective data stream indicated as the new event fail to satisfy at least one uncategorized event model.

2. The method according to claim 1, wherein the plurality of data sources comprises social media websites, online news websites, and blogs.

3. The method according to claim 1, wherein the templates define an event in terms of semantic roles of one or more entities.

4. The method according to claim 1, wherein the one or more ingestion modules comprises at least an optimized ingestion module configured to receive a raw data stream from a pre-defined list of data sources, and wherein the pre-defined list of data sources are selected by a common topic, type of data received, and format of publication.

5. The method according to claim 1, wherein the pre-processing reduces noise and redundancy of the plurality of raw data streams.

6. The method according to claim 1, further comprising transmitting, by the computer, the set of one or more raw data streams to a data processing module responsive to identifying the set of one or more raw data streams corresponding to the one or more ingestion modules.

7. The method according to claim 1, further comprising selecting, by the computer, one or more words in the plurality of raw data streams pertaining to the one or more events.

8. The method according to claim 7, further comprising assigning, by the computer, a weight to each of the one or more words.

9. The method according to claim 8, wherein the weight is assigned to each of the one or more words based on co-occurrence of each of these words with respect to the other words of the one or more words in the plurality of data streams.

10. The method according to claim 9, further comprising evaluating, by the computer, the weight assigned to each of the one or more words based on change in the co-occurrence number of each of these words with respect to the other words of the one or more words in the plurality of data streams.