Boris Robles-Slyusar

Feedback from colleagues and partners

Anonymized, merged into themes and grouped by month. So far it comes from classmates and my mentor. Notes from the data partners will join as they come in.

September 2026

  • Classmates

    How do I check the tool is right when it flags a review, and measure false positives?

    My take

    Classmates suggested testing on reviews already labeled real or fake, starting with a 2011 deceptive review corpus, but I have not committed to that. I need to work through the detection research, and have no answer yet on false positives.

  • Classmates

    Which 1 input, 1 outcome, 1 system, product category, and LLM platform do I start with?

    My take

    Reviews are the likeliest input, the outcome is shaping up as a re-ranked product list, and I have not locked in a system or category. LLM platforms may drop out of the first build, so that has no answer yet.

  • Classmates

    How does my tool avoid Fakespot's fate and keep up as GenAI reviews get harder to detect?

    My take

    Fakespot's shutdown is my case study. I think scraping cost at scale was a big factor. A classmate suggested labeled datasets instead of scraping. I have not settled the cost side or how the tool keeps up with GenAI reviews.

  • Classmates

    Is the tool about review trust or product quality, and what separates normal marketing from deception?

    My take

    I frame the tool around transparency and ranking credibility, not product quality, with the FTC Consumer Review Rule as my legal line for reviews. I have not decided where sponsored listings fall or how far into product quality to go.

  • Classmates

    Can shoppers spot fake reviews on their own, and would a verified-buyer filter help?

    My take

    No hard answer yet, but an estimated $770.7 billion in unwanted purchases in 2025 suggests shoppers miss fakes. Verified-buyer filters help only partly, since sellers pay people to buy and review. I have not decided how to use verified status.

  • Mentor

    Which marketing tactics beyond fake reviews should the tool cover, and how do I detect each?

    My take

    I have a working list of 10 tactics with detection signals for each, shown under Research. Reviews are the first input, and I have not decided which other tactics make the first build.

Next round

  • Open slot

    The next question lands here, from classmates, my mentor or a data partner.

    My take

    Written once the question is in. Yours is welcome: boris@optimizehub.net.

Partner on research

Have a question your data could answer?

If your team sits on marketing, sales or product data, with a question nobody has time to answer properly, that is the kind of partnership this research runs on. Free to participate. No commercial incentive. Just science. Reach me either way.

Message me on LinkedIn boris@optimizehub.net

Or read what partners get first.

Survey list

Send me a 1-line email

The subject "Notify me about the survey" is all it takes. You get 1 message when a survey opens, and 1 when the results are out. Nothing else.

Or open it in your mail app.