The present invention provides an apparatus for verifying the validity of a text and a method thereof. The apparatus includes: a text acquisition unit that obtains a plurality of self-replicating texts generated in a self-replicating manner by using texts for learning having been verified in advance for ethical or unethical properties; a dictionary-based discrimination unit that receives the self-proliferation text and searches for words similar to the profanity registered in the profanity dictionary obtained in advance from the approved self-replicating text and a pre-determined level or higher, and determines unethical properties of the self-replicating text; a learning model-based discrimination unit that receives the self-replicating text, vectorizes the text in units of words, and extracts sentence feature vectors from the vectorized words according to a previously learned pattern estimation method to determine the unethical properties of the self-replicating text; an original text-based discrimination unit that searches for text for learning most similar to the text for self-replicating, and determines the unethical properties of the text for self-replicating according to the label of the searched text for learning; and a discrimination result comparison unit that obtains the final discrimination result for the self-replicating text by combining the discrimination result of the unethical properties of the self-replicating text determined by each of the dictionary-based discrimination unit, the learning model-based discrimination unit, and the original text-based discrimination unit.
展开▼
