| CPC G06Q 50/265 (2013.01) [G06F 16/3344 (2019.01); G06F 16/35 (2019.01); G06F 16/383 (2019.01); G06F 40/30 (2020.01)] | 32 Claims |
|
1. A system for recommending related data based on similarity, the system comprising:
processor operably coupled to a memory comprising computer-readable instructions, wherein the processor is configured:
to collect unstructured open data together with meta data including information on the unstructured open data;
to receive the unstructured open data together with the meta data including the information on the unstructured open, and to generate and extract a predefined event from the open data and the meta data by using a pre-established event DB;
to construct an event data set by performing an event labeling operation for determining a similarity relation, on the basis of the event generated and extracted from the open data and the meta data;
to receive the event data set, on the basis of the event data set, an event vector by joining a relation between pieces of information on two different independent entities by using a word embedding technique;
to receive the event vector, and to establish, on the basis of the event vector, an event artificial intelligence learning model by performing learning for each of a plurality of the event vectors through a preset artificial intelligence (AI)-based machine learning method; and
to receive user open data to find similar data by extracting a feature vector of the user open data on the basis thereof, and to output the related data semantically similar to the extracted feature vector of the user open data using the event artificial intelligence learning model.
|