You may not realize it, but the apps on your smartphone may be collecting a lot of really personal data about you.
(Read more: Facebook's next target is your cursor )
Whether it's a recommendation, dating or other social app, companies are increasingly using machine-based learning to get to know users better so that they can provide an improved user experience and target ads more effectively.
But in order to do this, users must hand over access to other personal accounts, like Facebook or email.
And while many people are happy to give access to such accounts to reap the benefits of personalization, they probably don't realize just how much personal data they are actually sharing, and that can be dangerous, experts say.
(Read more: Why Online Privacy Is a Big Oxymoron)
"People should be very concerned about this, but it's a catch 22," said Mikael Berner, CEO of EasilyDo, a smart assistant application.
"There's no way we are going to survive these days without some automation tools, so people just get used to giving their information away, but they should be thinking about what data they are putting out there, starting with their social data."
Facebook also measures a user's behavior to determine how it ranks news in a user's newsfeed. The social network looks at not only who users are engaging with, but also what pages they interact with, how long a user scrolls through their newsfeed and how much time they spend on different pages to adjust their algos to display better content in the newsfeed.
When you think of social data, you probably think of the things you post on Facebook and Twitter, the pages you interact with on social platforms or the locations you check into on Four Square. But you actually share a lot more than you think simply by how you behave on different social platforms.
Much like how marketers use people's Web cookies to better target ads, apps you use also watch what you do on social platforms to improve personalization.
For example, the dating app Tinder uses information from your Facebook profile to find matches, but it also measures different data about you once you start using the app to help improve the user experience, including whom you are chatting with.
"We look at a variety of things, who your friends are, what your common interests are, where you went to college, different social factors and we implicitly look at who you are saying 'yes' or 'no' to and who you are having conversations with," said Sean Rad, CEO of Tinder.
Tinder uses the length and frequency of the conversation to determine if it should show you other users similar to the one you are conversing with.
"We are just looking at the frequency of the conversation, it's what we call the depth. We aren't looking at the semantics of the conversation," Rad said.
As for when a user says "yes" or "no" to a person's profile, Tinder looks at this information to determine trends so that it can adjust its algos to produce better match results.
(Read more: FTC suggests new guidelines for mobile apps)
Tinder isn't alone. Swoon, another dating app, uses similar tactics to pair users. In some cases, Swoon even measures a user's dwell time, or how long it takes for them to say "yes" or "no" to a match.
However, this metric is primarily used to spot unusual behavior on the platform, not to improve matches, said Johann Schleier-Smith, chief technology officer at Tagged, Swoon's parent company. He added that the company may use dwell time in the future as something to improve matches.
"We are really learning," Schleier-Smith said. "All of that data created by what users are doing when they use the app, it helps us learn more about them."
While it may seem creepy that apps watch behavioral patterns to better profile users and access user data from other social accounts, most consumers don't really care as long as they are satisfied with the product, Rad said.
"With Tinder, people begin getting matched within minutes of joining, which almost makes you not care about what you're sharing because you are getting value," Rad said. "The cost is justified,"
But the "cost" of what appears to be a free app is what many consumers may not understand, said Justin Brookman, director of consumer privacy for the Center for Democracy and Technology.
When consumers download and use a free app, they are actually paying with their data, And companies want to make money from users' data, even if they aren't sure just what their business model looks like yet, he said.
"With big data there's this idea that everyone out there wants to collect it, but they don't know what to do with it. They basically say, 'We have the right to collect this data on behalf of our client and we'll figure out what to do with it later,'" Brookman said.
"Unfortunately, a lot of these apps don't have a business model. The problem is they are all figuring out all their business model, and that's a concern going forward. You may be giving over the rights to certain information about you and you don't know what they will do with it."
Even if a company does seem to have a clear business model, that model might change if it's bought by a bigger company, like Google or Facebook, so it's difficult for consumers to keep up with how their information is being used.
(Read more: Google accelerates Glass rollout)
Because of consumer push back, larger social media companies have gotten better at being more clear with how they use people's information, though, Brookman said.
"These days you are seeing more and more consumers push back and companies are being forced to pivot a little bit," Brookman said.
"In general, companies realize people are more sensitive to privacy, and they don't want to wake up one day be on the cover of The Wall Street Journal. "
—By CNBC's Cadie Thompson. Follow her on Twitter .