Method for acquiring a parallel corpus, electronic device, and storage medium

An electronic device, comprising:

• at least one processor, and
• a memory, which is configured to store an executable instruction which, when executed by the at least one processor, causes the electronic device to implement:
• acquiring text data in a target document;
• performing language detection on the text data;
• in response to the text data comprising at least two languages, retaining the text data;
• detecting, in the retained text data, a distribution proportion of words in each of the at least two languages;
• in response to each of a distribution proportion of words in a first language and a distribution proportion of words in a second language being greater than a preset proportion threshold, retaining the detected text data, wherein the first language and the second language have highest distribution proportions of words;
• extracting, from the retained detected text data, sentences in the first language as a first sentence set and sentences in the second language as a second sentence set;
• performing sentence alignment on the first sentence set and the second sentence set;
• and
• determining sentence pairs with alignment degrees greater than a preset alignment degree threshold as the parallel corpus.

A non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements:

• acquiring text data in a target document;
• performing language detection on the text data;
• in response to the text data comprising at least two languages, retaining the text data;
• detecting, in the retained text data, a distribution proportion of words in each of the at least two languages;
• in response to each of a distribution proportion of words in a first language and a distribution proportion of words in a second language being greater than a preset proportion threshold, retaining the detected text data, wherein the first language and the second language have highest distribution proportions of words;
• extracting, from the retained detected text data, sentences in the first language as a first sentence set and sentences in the second language as a second sentence set;
• performing sentence alignment on the first sentence set and the second sentence set;
• and
• determining sentence pairs with alignment degrees greater than a preset alignment degree threshold as the parallel corpus.