Systems and method for generating event timelines using human language technology

A non-transitory computer readable storage medium having stored thereon a set of instructions for processing instructions of a software program that when executed by a computing device, cause the computing device to:

• receive a textual dataset;
• extract one or more time mentions from the textual dataset;
• extract one or more event trigger words from the textual dataset;
• extract one or more document creation times from the textual dataset;
• determine a relationship between an extracted event trigger word of the extracted one or more event trigger words from the textual dataset and the extracted time mention of the one or more extracted time mentions from the textual dataset, wherein determining the relationship between the extracted event trigger word from the textual dataset and the extracted time mention from the textual dataset includes training a machine learning classifier to categorize the relationship between the extracted event trigger word and the extracted time mentions;
• determine a relationship between an extracted event trigger word of the extracted one or more event trigger words from the textual dataset and an extracted document creation time of the one or more extracted document creation times from the textual dataset, wherein determining a relationship between the extracted event trigger word from the textual dataset and the extracted document creation time from the textual dataset includes training a machine learning classifier to categorize the relationship between the extracted event trigger word and the extracted document creation time;
• and
• generate a timeline, wherein the timeline is based on the one or more extracted event trigger words, the one or more extracted time mentions, the determined relationship between the extracted event trigger word from the textual dataset and the extracted time mention from the textual dataset, and the determined relationship between the extracted event trigger word from the textual dataset and the extracted document creation time from the textual dataset.

An electronic device, the device comprising:

• a memory;
• and
• one or processors configured to:
• receive a textual dataset;
• extract one or more time mentions from the textual dataset;
• extract one or more event trigger words from the textual dataset;
• extract one or more document creation times from the textual dataset;
• determine a relationship between an extracted event trigger word of the extracted one or more event trigger words from the textual dataset and the extracted time mention of the one or more extracted time mentions from the textual dataset, wherein determining the relationship between the extracted event trigger word from the textual dataset and the extracted time mention from the textual dataset includes training a machine learning classifier to categorize the relationship between the extracted event trigger word and the extracted time mentions;
• determine a relationship between an extracted event trigger word of the extracted one or more event trigger words from the textual dataset and an extracted document creation time of the one or more extracted document creation times from the textual dataset, wherein determining a relationship between the extracted event trigger word from the textual dataset and the extracted document creation time from the textual dataset includes training a machine learning classifier to categorize the relationship between the extracted event trigger word and the extracted document creation time;
• and
• generate a timeline, wherein the timeline is based on the one or more extracted event trigger words, the one or more extracted time mentions, the determined relationship between the extracted event trigger word from the textual dataset and the extracted time mention from the textual dataset, and the determined relationship between the extracted event trigger word from the textual dataset and the extracted document creation time from the textual dataset.