Limited Offer Get 25% off — use code BESTW25
No AI No Plagiarism On-Time Delivery Free Revisions
Claim Now
Prices start at $5 per page · plagiarism-free papers · free revisions for 14 days 24/7 support: support@bestwriters.com
Uncategorized

initial thought of comparing the performance

  • Posted
  • Authoradmin
  • Length467 words
  • Reading time3 min
Write My Assignment FREE

According to the discussion between my supervisor and me, we agree that my initial thought of comparing the performance of existing fraud detection tools on electronic product reviews is hard to work on well as we cannot find golden standard corpus in the domain of electronic product that is already been labelled as fake review or reliable review. At that point, I found a corpus on Kaggle and my supervisor liked it, so he provide me with another preliminary topic ‘A comparison study of sentiment analysis tools in the context of user reviews based on explicit rating measures’ based on that mobile phone corpus (https://www.kaggle.com/PromptCloudHQ/amazon-reviews-unlocked-mobile-phones). He wanted me to look into reviewing sentiment analysis tools based on the Kaggle corpus or similar. Your gold standard is the rating, so a review with 4-5 stars is positive, 3 is neutral and 1-2 is negative. Then you could test the performance of the tools against the rating to see whether the detection of the sentiment agrees with the rating. So last meeting he asked me to find out the 2-3 existing sentiment analysis tools (he also provide me with a tool named GATE https://gate.ac.uk/) that I want to investigate, and find one or two more corpus with rating and reviews similar with the mobile phone corpus (He suggested the restaurant reviews with star rating). Therefore, I can compare them two by two (two tools and two corpus) and see their performance and discuss according to the result. Also in the last email, he said if I cannot find the corpus with rating, I can also use the labelled corpus (e.g. https://nlp.stanford.edu/sentiment/code.html).

But I am still confused about what is the aim of this type of comparative research and what question I want to answer in this dissertation topic. So, I want you to help to adjust this topic to be more meaningful and feasible. But if you think that this topic makes little sense, the research domain of my supervisor is NLP and particularly negation detection (He published an article about Negation detection on archeology and one of his suggested dissertation area for us is Negation Detection approaches in the humanities domain), could you please find a feasible topic in this area.

At the same time, I also have a concern that these sentiment analysis tools are designed with multi-domain issues in mind (it can also be said that the tools are not domain-specific), so if I just take the corpus of electronic products and restaurants into account and conduct experiments to test their accuracy, what if the final results are similar and there are no useful discoveries?

The post initial thought of comparing the performance appeared first on My Assignment Online.

Plagiarism Free Assignment Help

Expert Help With This Assignment — On Your Terms

✓ Native UK, USA & Australia writers ✓ Deadline from 3 hours ✓ 100% Plagiarism-Free — Turnitin included ✓ Unlimited free revisions ✓ Free to submit — compare quotes

admin

Academic writer and editor at BestWriters.

Get your paper off your plate tonight

Post your instructions for free. Prices start at $5 per page.

Place an order