Perception capture system

A computing system comprising:

• a memory that stores instructions;
• and
• one or more processors configured by the instructions to perform operations comprising:
• training a predefined number of machine learning models using the same hyper-parameters to configure each of the predefined number of machine learning models to generate a vector representation for a given word;
• inputting a first word of a plurality of words in social media data from a plurality of social network platforms in a predefined window of time into each machine learning model of the predefined number of machine learning models to generate a vector representation for the first word from each machine learning model;
• generating a set of vector representations for the first word comprising all vector representations for the first word output from the predefined number of machine learning models;
• performing operations, until there are no more words of the plurality of words left to input, comprising:
• inputting a next word of the plurality of words into each machine learning model of the predefined number of machine learning models to generate a vector representation for the next word from each machine learning model;
• and
• generating a set of vector representations for the next word comprising all vector representations for the next word output from the predefined number of machine learning models;
• generating a set of similarity scores for each set of vector representations for each word compared to a given name;
• generating a perception score for each word based on the set of similarity scores for the set of vector representations for each word;
• identifying top concepts based on the perception score for each word;
• and
• comparing the top concepts to previously determined concepts in previous time periods.

A nontransitory computer-readable medium comprising instructions stored thereon that are executable by at least one processor to cause a computing system to perform operations comprising:

• training a predefined number of machine learning models using the same hyper-parameters to configure each of the predefined number of machine learning models to generate a vector representation for a given word;
• inputting a first word of a plurality of words in social media data from a plurality of social network platforms in a predefined window of time into each machine learning model of the predefined number of machine learning models to generate a vector representation for the first word from each machine learning model;
• generating a set of vector representations for the first word comprising all vector representations for the first word output from the predefined number of machine learning models;
• performing operations, until there are no more words of the plurality of words left to input, comprising:
• inputting a next word of the plurality of words into each machine learning model of the predefined number of machine learning models to generate a vector representation for the next word from each machine learning model;
• and
• generating a set of vector representations for the next word comprising all vector representations for the next word output from the predefined number of machine learning models;
• generating a set of similarity scores for each set of vector representations for each word compared to a given name;
• generating a perception score for each word based on the set of similarity scores for the set of vector representations for each word;
• identifying top concepts based on the perception score for each word;
• and
• comparing the top concepts to previously determined concepts in previous time periods.