| CPC G06Q 10/06398 (2013.01) [G06Q 10/06393 (2013.01); A63B 24/0062 (2013.01); G09B 19/0038 (2013.01)] | 16 Claims |
|
1. A system for generation and traversal of a skill representation graph using machine learning, the system comprising:
a computing device, the computing device configured to:
receive a plurality of data of a plurality of individuals, wherein the plurality of data comprises a plurality of individual skill levels corresponding to a common skill of a plurality of common skills, and wherein the plurality of data combines a dynamic network of users who train together and a training history for each user;
determine a relative skill level of the plurality of individuals from the plurality of data, wherein the relative skill level includes an achievement level determined by:
calculating a mean difficulty per individual per skill of the plurality of data of the plurality of individuals; and
calculating the achievement level as a percentile of the mean difficulty, among all individuals and the relative skill;
generate a skill representation graph representing a plurality of skill interrelations, each interrelation of the plurality of skill interrelations representing a degree of difficulty in acquiring a second skill after acquiring a first skill, as a function of the relative skill level, wherein generating the graph further comprises:
generating a plurality of nodes wherein each node represents a skill;
generating a plurality of interconnections wherein each interconnection represents a process and/or path to master a subsequent skill of a first skill;
generating the plurality of interrelations as a function of the at least a plurality of data and a neural network, wherein generating the plurality of interrelations further comprises:
training, using the plurality of data, the neural network to output a plurality of embeddings representing skills associated with nodes, wherein the neural network further comprises:
an input layer;
at least a hidden layer; and
an output layer, the output layer configured to:
compare pairs of adjacent nodes to at least a threshold according to a pairwise similarity test; and
remove interconnections, of the plurality of interconnections, between pairs where the similarity test does not meet the threshold;
outputting, using the neural network and the plurality of nodes, the plurality of embeddings;
determining a plurality of distances between the plurality of embeddings using a distance measure; and
generating the plurality of interrelations as a function of the plurality of distances; and
assembling the graph using the plurality of interrelations, wherein assembling the graph further comprises:
representing the plurality of interconnections as edges between the plurality of nodes representing skills; and
representing distances between embeddings corresponding to pairs of connected skills as lengths of edges connecting corresponding nodes;
determine at least a goal skill of the plurality of common skills by recommending to the user a third skill and a fourth skill;
wherein the third skill is determined as a function of which skills the user has acquired, which skills are interconnected to at least one skill the user has acquired, and which skills have higher average difficulty than the average difficulty of all skills the user has acquired;
wherein the fourth skill is determined as a function of a lowest achievement level of a skill of the user; and
determine at least one improvement activity to achieve the at least a goal skill; and
a user device, wherein the user device is configured to:
display the skill representation graph and at least one indicia of the user's current state of progress of an improvement activity in achieving the at least a goal skill as a function of the computing device.
|