Practice Test 1: Sentiment Classification Technique Families
Practice test on machine learning, lexicon-based, and hybrid classification techniques from the survey.
6 questions · pass mark 50% · try 5 free
1 Which set correctly names the three main classes of sentiment analysis techniques identified in the survey?
- Machine learning, lexicon-based, and hybrid
- Parsing, translation, and summarization
- Clustering, stemming, and indexing
- Neural, symbolic, and probabilistic only
The survey groups classification techniques into machine learning, lexicon-based, and hybrid methods.
2 Which of the following are machine learning subgroups mentioned in the survey?
Select all that apply
- Supervised
- Semi-supervised
- Unsupervised
- Dictionary-based
The survey divides machine learning approaches into supervised, semi-supervised, and unsupervised methods. Dictionary-based belongs to lexicon-based approaches.
3 The survey describes supervised learning as generally having good accuracy but high cost.
- True
- False
This is true: supervised learning is presented as accurate but costly because of labeled data and training requirements.
4 What is a major advantage of unsupervised learning according to the survey?
- It always produces the highest accuracy
- It does not require much human participation
- It eliminates all domain effects
- It guarantees stable results
The survey notes reduced human participation as a key strength of unsupervised approaches.
5 Lexicon-based methods require labeled training samples in the same way supervised learning does.
- True
- False
This is false: lexicon-based methods are noted as not requiring labeled training samples.
1 more question in this test
Take the full test in MagicalTutor with instant marking, an explanation for every answer, and a score tracked against your weak topics.
Take the full test