Online movie rental portal Netflix is looking to devise a way to make its movie-recommendation service a little more accurate, and it's turning to customers for ideas. The internet rental service said this week that it has created the Netflix Prize - a $1 million award - to the first person who can achieve greater accuracy goals when recommending movies based on personal preferences.

The company also made available to contestants 100 million anonymous movie ratings ranging from one to five stars.

According to the company, recommendation systems can help people choose taste-based products by comparing their preferences with similar tastes of other people. Recommendation systems are particularly useful to consumers of movies, books, movies, clothing and entertainment, the company said.

The threshold required to win the Netflix Prize is a 10 percent improvement in accuracy over the company's current recommendation system. If no one wins the grand prize this year, the company said it will award a $50,000 progress prize to whoever makes the most significant advancement toward the goal and will award a progress prize annually until someone wins the grand prize.

Netflix said it will publish a detailed description of the winning approach for the benefit of companies, entrepreneurs and academics.