DESC:TIME-SERIES ANALYSIS BASED ON WORLD EVENT DERIVED FROM UNSTRUCTURED CONTENT ,CLAIMS:1. A method for analyzing time-series data based on world events derived from unstructured content, the method comprising:
obtaining event information corresponding to at least one world event from the unstructured content obtained from a plurality of data sources, wherein the event information includes at least time of occurrence of the at least one world event, time of termination of the at least one world event, at least one entity associated with the at least one world event, and a pre-assigned event class;
retrieving, time-series data pertaining to the at least one entity associated with the at least one world event, from a time-series data repository; and
aligning, the at least one world event with at least one time-series event based on the event information and the time-series data, to identify at least one pattern indicative of a cause-effect relationship between the at least one world event and the at least one time-series event.
2. The method as claimed in claim 1, wherein the method further comprising:
extracting a plurality of world events from the unstructured content, wherein the plurality of world events include the at least one world event; and
assigning a unique event identifier (ID) to each of the plurality of world events.
3. The method as claimed in claim 2 wherein the extracting comprises:
obtaining meta-data from the unstructured content, wherein the meta-data is indicative of attributes of the unstructured content;
de-duplicating the unstructured content; and
grouping similar unstructured content into a plurality of groups, wherein the unstructured content present in each of the plurality of groups corresponds to a world event.
4. The method as claimed in claim 3, wherein the extracting further comprises:
tokenizing the unstructured content into a plurality of sentences and words;
tagging each word of the plurality of sentences with post script tags;
extracting one or more phrases from each of the plurality of sentences;
identifying verbs from each of the plurality of sentences based on the tagging and using a verb lexicon;
classifying verbs into a plurality of verb-classes based on a verb lexicon, wherein each of the verb-classes represents syntactically and semantically correlated verbs;
identifying entities in each of the plurality of sentences based on a named entity recognization technique; and
extracting relations involving at least one entity in each of the plurality of sentences based on a relation extraction technique.
5. The method as claimed in claim 1, wherein the method further comprises indexing the event information corresponding to the at least one world event with the time series data of the at least one time-series event and the at least one pattern.
6. The method as claimed in claim 1, wherein the method further comprises performing a data analytics activity based on the at least one pattern, the data analytics activity includes one of predicting at least one future time-series event, and evaluating an operational risk based on an impact of the at least one world event on the at least one time series event.
7. The method as claimed in claim 1, wherein the method further comprises linking the at least one world event with at least one pointer pointing to an original full text document from which the at least one world event is extracted.
8. A data analytics system (102) comprising
a processor (202);
an analysis module (108) coupled to the processor (202), to:
obtain event information corresponding to at least one world event (232) from unstructured content (224) stored in an unstructured data repository (222), wherein the event information includes at least time of occurrence of the at least one world event (232), time of termination of the at least one world event (232), and at least one entity associated with the at least one world event (232);
receive, time-series data (228) pertaining to the at least one entity associated with the at least one world event (232), from a time-series data repository (226); and
align, the at least one world event (232) with at least one time-series event based on the event information and the time-series data (228), to identify at least one pattern indicative of a cause-effect relationship between the at least one world event and the at least one time-series event.
9. The data analytics system (102) as claimed in claim 8, wherein the data analytics system (102) further comprises an indexing module (218) to index the event information corresponding to the at least one world event with the time series data of the at least one time-series event and the at least one identified pattern to perform a data analytics activity.
10. The data analytics system (102) as claimed in claim 8, wherein the analysis module (108) performs a data analytics activity based on the at least one pattern, wherein the data analytics activity is one of prediction of at least one future time-series event based on the at least one pattern, and evaluation of an operational risk based on an impact of the at least one world event on the at least one time series event.
11. A non-transitory computer-readable medium having embodied thereon a computer program for executing a method comprising:
obtaining event information corresponding to at least one world event from unstructured content obtained from a plurality of data sources, wherein the event information includes at least time of occurrence of the at least one world event, time of termination of the at least one world event, at least one entity associated with the at least one world event, and a pre-assigned event class;
retrieving, time-series data pertaining to the at least one entity associated with the at least one world event, from a time-series data repository; and

aligning, the at least one world event with at least one time-series event based on the event information and the time-series data, to identify at least one pattern indicative of a cause-effect relationship between the at least one world event and the at least one time-series event.