Automatic Social Media Content Timeline Summarization Method and Apparatus

A non-transitory computer-readable storage medium tangibly encoded with computer-executable instructions, that when executed by a processor associated with a computing device, performs a method comprising:

• receiving a request from a social media service user for information about an entity;
• identifying a plurality of social media content items, the social media content item identification comprising searching a number of social media content item data stores for the plurality of social media content items containing information about the entity;
• determining, using each social media content item of the plurality, an episode timeline comprising a number of episodes, the episode timeline determination comprising modeling, using statistical machine modeling, content of the plurality of social media content items in combination with temporal information of the plurality of social media content items, the timeline determination further comprising assigning a number of social media content items of the plurality to each episode of the number;
• determining, for an episode, a ranking of the number of social media content items assigned to the episode, the rank determination comprising determining a set of features for each social media content item of the number and using statistical machine modeling to rank the social media content items of the number;
• automatically selecting, for the episode, a predetermined number of top-ranked social media content items from the number assigned to the episode;
• and
• automatically creating a timeline summarization of the plurality of social media content items, the timeline summarization comprising a number of entries corresponding to the number of episodes of the episode timeline, an entry comprising the content of the predetermined number of top-ranked social media content items automatically selected for a corresponding episode of the number of episodes.

A computing device comprising:

• a processor;
• a non-transitory storage medium for tangibly storing thereon program logic for execution by the processor, the program logic comprising:
• receiving logic executed by the processor for receiving a request from a social media service user for information about an entity;
• identifying logic executed by the processor for identifying a plurality of social media content items, the social media content item identification comprising searching a number of social media content item data stores for the plurality of social media content items containing information about the entity;
• determining logic executed by the processor for determining, using each social media content item of the plurality, an episode timeline comprising a number of episodes, the episode timeline determination comprising modeling, using statistical machine modeling, content of the plurality of social media content items in combination with temporal information of the plurality of social media content items, the timeline determination further comprising assigning a number of social media content items of the plurality to each episode of the number;
• determining logic executed by the processor for determining, for an episode, a ranking of the number of social media content items assigned to the episode, the rank determination comprising determining a set of features for each social media content item of the number and using statistical machine modeling to rank the social media content items of the number;
• selecting logic executed by the processor for automatically selecting, for the episode, a predetermined number of top-ranked social media content items from the number assigned to the episode;
• and
• creating logic executed by the processor for automatically creating a timeline summarization of the plurality of social media content items, the timeline summarization comprising a number of entries corresponding to the number of episodes of the episode timeline, an entry comprising the content of the predetermined number of top-ranked social media content items automatically selected for a corresponding episode of the number of episodes.