There is a lot of controversy on copycats (repetition of the original idea, without making changes). Is it worth repeating Western counterparts, if so, is it worth it to veil and at least modify the interface. There is no consensus on this. We decided to repeat one of the western startups.
Many have probably heard about the
Aardvark service, and about its subsequent purchase by Google. We did not immediately repeat it. We waited some time to see if Google would adapt it. When we realized that no steps were being taken in this direction, we decided to try ourselves in a social search, especially since we ourselves regularly used Aardvark and faced the problem of the lack of support for the Russian language and the lack of a sufficient number of users from the Russian-speaking region.

This is how
huti.ru appeared. It is based on the same idea that Aardvark has - a social search in which you can ask a question, and the service itself determines the subject of the question and looks for suitable people who can answer it. To start work you need to register and specify your expert areas. Next you need to add yourself in the contacts of the bot, which you can at any time ask an interesting question.
The algorithm that we laid at Houthy is:
1. The user asks a question.
2. The system selects keywords based on frequency characteristics like
TFIDF .
3. Next, the system builds a list of topics with weights that correspond to the selected words and phrases. Knowledge about the topics and words that describe them were obtained on the basis of public information which includes Google answers, Yandex answers, Wikipedia,
dmoz .
4. Next, the system searches for users whose topics most closely match the topics and weights of the question.
5. This weight is multiplied by the characteristics of the user - how often he responds, how often his answers suit those who asked, how much their age corresponds, etc.
6. The system polls the users one by one until someone answers and until the questioner responds to the answer. Thus, the smallest number of people is involved in this process and (by design) the anxiety of the system’s participants is minimal.
7. The question closes as soon as a sufficient number of satisfactory answers is received or the maximum of people who have seen the question is exceeded. This way we ensure the quick closure of irrelevant questions, as well as those questions, the answer to which has already been received.
')

The main problem encountered in the initial stages and which is still relevant is the insufficient core of users and topics. And since the system to any question in any case finds people who, in its opinion, can answer it, there are cases when the found user is thematically actually far enough away from the topic of the question. We hope to solve this problem with an increase in the number of users and, accordingly, an expansion of the thematic base.
To date, the statistics are as follows:
Users: 600
Questions: 1700
4300 replies
Average speed of receiving an answer to a question: 30 minutes
Supported messengers: GTalk, Jabber, ICQ (not always online). There was another bot for Facebook chat, but it was banned, apparently suspected something was wrong.
We will be happy if the service will help someone quickly find answers to questions, or if you are ready to share your knowledge with others.
To register, you can use the site. After that, you need to specify your Messenger contact on the appropriate page and then add the bot Huthi to your contacts (Huty himself will not knock on you even after he knows your contact).
Or go faster in a huti-way — add a new contact to your messenger (the list of contacts here is
huti.ru/staticpages/about_better ), send him the text habr_huti_invitation, and go through a quick registration in question-answer mode.
Update: Because of the site’s flaws, we recommend that users of GTalk and Jabber still register and communicate with Huty using the messenger. How to log in on the site for those who registered via the instant messenger, see
p.5 -
huti.ru/staticpages/about_help