Top restaurants: watkins grill makoto sushi el paso restaurants clinton sc nicolinis restaurant boardman oh el campesino mcmurray asher dairy bar menu perfect perks washington nc thai garden oceanside j story cafe chicago thai house cuisine hillsboro or 609 se ankeny st, suite c portland, or 97214 ny pizza alachua fl dominos tyler tx gentry starbucks navajo road pizza hut west sunshine springfield mo hungry hunter patterson ga aries colony hall southfield mi michigan coni island saginaw mi menu pizza hut on jefferson avenue pizza hut 100th and wadsworth salsa verde restaurant yorkville il tai sushi house mckinney tx vittoria restaurant powell ohio menu pizza hut rendon burleson tx motown coney island woodward menu blue ocean sushi & asian grill lewisville tx
Frequently Asked Questions
- What is t-jikoku?
t-jikoku is a large database of restaurants that collects rating data about those restaurants. We then use collaborative filtering to make predictions about restaurants you might like that you haven't visited yet (or at least rated).
- Why is t-jikoku different from other Yellow Page sites?
t-jikoku offers much more than just a big directory of restaurants. In addition to making educated guesses about new restaurants you might like we can also combine multiple user profiles to find the best restaurants for those users. So for example if two people are going on a date we can offer a list of restaurants that they might both enjoy. We also allow our users to create profile pages to share their favorite restaurants with anyone.
- Why not just have reviews? Why is collaborative filtering so important?
It can be difficult to easily determine if a reviewer has similar tastes as you do. Perhaps the reviewer loves a particular type of atmosphere or food that you can't stand. What is that value of a good review from this person to you?
With t-jikoku.info's collaborative filtering algorithm we will detect that this reviewer does not match your tastes and we will throw out their opinion (just for you, because they are probably relevant for other people).
- What is collaborative filtering?
Collaborative filtering is a method that collects a large amount of user preference data (in our case restaurant ratings), and uses that data to infer what other users might think about a restaurant they've not yet rated. The more data that is entered the better the system becomes for everyone because the prediction algorithms will understand each user better.
For example, if Jim, Sally, and Frank all like restaurants A, B, C and D, but don't like E and Bill has indicated that he likes restaurants B, C, and D, but not E then it is likely that Bill will also like restaurant A. Is this method perfect? No way, but it can give great suggestions if you are looking to try something new. Of course if the restaurant suggested is not up to par, please leave a recommendation so that our prediction algorithm can take that into consideration the next time you use the site.
- So did you invent collaborative filtering?
- So you are using my personal information for predictions?
No, actually all predictions are made solely on the basis of ratings and some other restaurant listing data. In fact the only personal data outside of standard web statistics that we collect is used to generate driving directions to save you some time typing. This information is completely optional.
- Is this free?
Yes! Our goal is to have the most comprehensive directory of restaurants available and we believe that allowing the t-jikoku.infomunity to add, update, or even delete out of date listings for free is the best means to accomplish this. In order to make any changes or get predictions for restaurants we do require that you sign up for a free account with us first.
- How many restaurants do you currently have in your directory?
The number of restaurants is continuing to increase, however we currently have just over 700,000 restaurants.
- I am able to contribute a large number of restaurants, however I have them in electronic format and don't want to type them into your form, what can I do?
First of all, we love doing this, so don't hesitate to ask! Please contact us and we will find a way to integrate your data into our site.
- Why are you doing this?
After briefly entertaining the idea of competing for the Netflix Prize we instead decided to take all of our new knowledge about collaborative filtering and apply it to an area other than movie recommendations. After some investigation we found that nobody else was doing this for restaurants in any major capacity, and therefore this might be a really excellent tool for the online community.