Arabic: Mentions extraction and sentiment precision

For this project, you will have to analyze two files: * Coverage file: you will have to manually extract mentions from 200 reviews belonging to the hospitality industry (hotels and restaurants). You will receive a file with 200 reviews in the specified language and an explanation of the main doubts. * Sentiment precision: you will have to validate the sentiment detected by the system (positive/negative/neutral) from 1000 mentions (short snippets of text of around 6 words). NOTE: In order to ensure the expected quality of the files, the accepted margin of error of the Sentiment precision file is of around 5% based on previous analysis. The payment will not be released until the expected margin of error is ≤ 5%