Facebook founder Zuckerberg proposed to do a search, a lot of people have expressed great interest, because Zuckerberg said the future of search can be tailored to the temptation is very large. In January 15, 2013, the Facebook graph search search tools available, vector customized search first step.
first, graph search is to provide high quality information for users, with great needs of users. User activity on Facebook, the greater the amount of Facebook amount of data is greater, the user relationship is more abundant, the organization search results can be more accurate and full. Such as the above search "2010 high school reunion photos, only people took photos to share out the only photos at all, even the photos are not that cannot search, and upload more photos, more rich content. In addition, the more people share, contact between students more closely, the dissemination scope is wide. Moreover, additional information more every photo, such as name, background, theme, the more able to meet the search needs more content, and depth is better. And so on, users around these photos happen rich and colorful, the search results are more rich and colorful, this circle of search is more valuable.
from this simple analysis, we can see some of the charm of social search, leaving a lot of vision and imagination, but also can be found that the challenge of social search. From the above example, at least we can dig out the two major challenges facing social search:
is a customized search, everyone can get their most needed information according to their own needs, but also can be said to be alive and only a few useful information. The realization of all this, Facebook is based on powerful user relationship. This is the graph search user relationship organization the lifeblood of personalized search results.
found a "2010 high school reunion photos", search to the course is the senior class of the class of 2010 party photos, this search only for yourself or your class is useful to people, other people don’t know how it happened, there is no interest. But the search terms in the Google search box, perhaps Google will return tens of thousands of mark to you.
on the contrary, if a circle of user activity is very small, the relationship is quite loose, the Facebook is also very difficult to organize this circle search content. For example, the circle of people to see such good stuff is not "like", or "share", graph search might not know the content of these recommended to who may simply follow the traditional keyword matching search results are given. The keyword matching the search result is very rough.