by Vas Mylko, Roman Bilusiak, Randal Olson
We started full-time on the 1st of April 2018 with the mission to build Curiosio — the smartest travel guide. We are recreating the mental process of a traveler, building machine empathy that is feeling the traveler. Curiosio is a high-tech startup.
The mission is hard but possible according to the progress we have made. We already released 12 beta versions (11 public, 1 private) and API. Currently, we are working on the public beta13 that will have cool themes.
Geographical Coverage
As of today, we released 17+ countries: Argentina, Brazil, Canada, Chile, Mexico, United States, France (with Monaco), Germany, Italy (with San Marino and Vatican), United Kingdom, India, Pakistan, Sri Lanka, Turkey, Australia, New Zealand, South Africa (with Lesotho and Eswatini). 2 countries are being tested in the Lab: Peru, Bolivia.
For the released 17 countries Curiosio knows ~210K census places (cities, towns, villages) and parks, ~340K points of interest, ~22M kilometers (~14M miles) of roads.
Team & Science
The team has been extended to three. The third geek is Randal Olson. The one you [all] know for his record-breaking routes through the US national parks, mini-breweries, state capitals, etc. Randal is a community leader of the Data Is Beautiful subreddit /r/dataisbeautiful. Randal loves to travel and knows many cool places by heart (like Thor’s Well).
Dr. Randal S. Olson’s research specialties are Artificial Intelligence (Machine Learning, Evolutionary Computation), Data Science (Large-Scale Data Analysis, Statistics, Data Visualization), Biology (Artificial Life, Neuroscience, Evolution of Intelligence). He is renown leader of the Genetic and Evolutionary Computation Conference (aka GECCO): GECCO 2017, GECCO 2016 best paper Evolutionary ML, GECCO 2013 best paper Artificial Life.
Besides creation of the unprecedented artificial smartness we are going to break Guinness World Records!
Science & Education
Vas mentors two students who are implementing relevant [to Curiosio] researches.
The first student is working on the GeoNER aka Geographic Named Entity Recognition. The idea is to crawl all travel stories on the open web and parse the itineraries from them. Make the interactive versions of those stories for reuse with customization (by points, duration, budget, etc). We are going to parse National Geographic articles in the scope of the diploma. Then we will see.
The second student is working on the multi-day trip planned within a single destination that creates daily programmes automatically for the defined number of days and budget. Each daily programme will consist of See & Do for the best experience and according to the geographical location of the stay. It’s multi-objective multi-modal optimization of NPC computational complexity. We are going to pilot in Lviv in the scope of the diploma.
Plans
Tighter and deeper integration with Atlas Obscura. We are building more for travelers, not tourists. Travelers love Atlas Obscura, Geek Atlas; tourists love Tripadvisor…
Integration with Tesla. Then Apple. Apple Car and Tesla Phone in 2022:)
Stay tuned and follow your curiosity.
