Don’t stereotype me

Don’t stereotype me

Collaboration vs. Personalization Approach
Personalization is all about people experiencing personalization; it cannot be answered on statistical model of future behavior prediction alone, but as something that one has lived through.
Personalized experience is a subjective observation in everybody’s life and it’s always personal feeling of emotions and sensations as opposed to thinking.
Now days, because of monetization reasons corporate are trying to answer the question of how to better serve their customers in order to increase revenues.
Matching between services and an online person is the new killer feature on the internet today.
How can we force the digital box to think out of the box and start the new digital life spine?

In order to implement personalization mechanism the system should predict in some level what will catch the user’s attention at any giving time.
On the Matrix movie, the Oracle told Neo not to worry about the vase, as Neo turns to look for a vase, and as he does, he knocks over a vase of flowers, which shatters on the floor. Then the Oracle asks Neo “would you still have broken it if I hadn’t said anything?”
Did the Oracle predict what’s going to happen or maybe she manipulates Neo and cause him to break the vase and does it really matter? Either way, Neo experienced this situation as prophecy.

Our individuality is unique
Recommendation system needs to predict person’s choice and interest and also to make a persona create new behavior. Personalization is happening when the content adapts itself based on the person’s profile, and provides something new, different, and possibly unexpected, is this a paradox? Does the human’s minds so predictive?
In order to provide a working recommendation system, there are two major methods;

- Collaboration Filtering (CF); recommendation agents employ prediction algorithms to provide users with items that match their interests based on their virtual community.
- Profile-based recommendation; history as an indication of user’s current or future desires.
Both methods have pros and cons and the combination of both will be a good predictor for any recommendation system.

Collaboration filtering, predict a person interest by analyzing the behavior of the virtual tribe he belongs to at any given moment. By being one of many alike, we all feel special but do we act the same or we’re all individuals?

Behavioral patterns
Can our past behavior predict our future behavior?
If you will ask a person how he will react in a certain situation, although he can’t really know, his answer will be based on his past, based on his intention and past behavior-future behavior relations.
Previous experience is our best way to answer any question that should predict our actions in the future, we can imagine it, we can vision our self’s based on our beliefs, faith and hopes.
But is previous behavior is the best predictor of our future interest? In some extant probably yes, but as our intention is dynamically changing the personalization system that based only on our past behavior probably won’t predict what will be our next click…

As I mentioned at the beginning, personalized experience is a subjective observation and it’s always, always a personal feeling of emotions, so I must state that I don’t feel well when the big news portals stereotype me based on my virtual tribe.

By: Sol Tzvi,CEO
- Back to: GENiEO Personal Homepage

About the Author

Miss Tzvi has held a variety of R&D and marketing positions in leading hi-tech industry companies in the past 12 years. She held several top managerial positions at Microsoft, both at the Israeli branch and at Microsoft headquarters in Seattle. Sol led Microsoft’s strategic marketing approach for information security and privacy in North America. Prior to founding Genieo Innovation, Miss Tzvi acted as a strategic consultant for private equity investments in several startup companies. She has also published three professional books and many articles.