Systems and methods for identification and classification of social media

A social media analytics method for improving the accuracy of the identification and classification of a social media posting, using one or more processors, the method comprising:

• collecting, by a social media analytics server from one or more social media data sources, image-based data from one or more social media data sources associated with a social media posting;
• collecting, by the social media analytics server from the one or more social media data sources, text-based data from one or more social media data sources associated with the social media posting;
• collecting, by the social media analytics server from the one or more social media data sources, reaction data from one or more social media data sources associated with the social media posting;
• extracting a first set of event related data from each of the image-based data, the text-based data, and the reaction data of the social media posting, wherein each of the image-based data, the text-based data, and the reaction data together is indicative of a given life event associated with a social media user who posted the social media posting;
• training a machine-learning based predictive model on a second set of event related data comprising each of (1) image-based data, (2) text-based data, and the (3) reaction data, the second set of event related data representative of life events of social media postings;
• inputting each of the (1) image-based data, (2) the text-based data, and the (3) reaction data of the first set of event related data;
• determining one or more life event classifications based upon the first set of event related data, wherein the determining includes the machine-learning based predictive model analyzing at least a portion of the image-based data comprising one or more extracted human body portions having distinct human feature distance values indicative of the given life event;
• and
• triggering an activity based upon the one or more life event classifications, wherein the triggered activity includes sending a communication regarding the one or more life events to a social media user device of the social media user, wherein the communication is displayed on the social media user device.

A tangible, non-transitory computer-readable medium storing instructions for improving the accuracy of the identification and classification of a social media posting, that, when executed by one or more processors of a computer system, cause the computer system to:

• collect, by a social media analytics server from one or more social media data sources, image-based data from one or more social media data sources associated with a social media posting;
• collect, by the social media analytics server from the one or more social media data sources, text-based data from one or more social media data sources associated with the social media posting;
• collect, by the social media analytics server from the one or more social media data sources, reaction data from one or more social media data sources associated with the social media posting;
• extract a first set of event related data from each of the image-based data, the text-based data, and the reaction data of the social media posting, wherein each of the image-based data, the text-based data, and the reaction data together is indicative of a given life event associated with a social media user who posted the social media posting;
• train a machine-learning based predictive model on a second set of event related data comprising each of (1) image-based data, (2) text-based data, and (3) reaction data, the second set of event related data representative of life events of social media postings;
• input each of the (1) image-based data, (2) the text-based data, and the (3) reaction data of the first set of event related data;
• determine one or more life event classifications based upon the first set of event related data, wherein the determination includes the machine-learning based predictive model analyzing at least a portion of the image-based data comprising one or more extracted human body portions having distinct human feature distance values indicative of the given life event;
• and
• trigger an activity based upon the one or more life event classifications, wherein the triggered activity includes sending a communication regarding the one or more life events to a social media user device of the social media user, wherein the communication is displayed on the social media user device.