An Indian-origin scientist based in the US has developed a computer programme that can help film producers determine whether their upcoming flicks would be a hit or a miss at the box office.
Ramesh Sharda, an information scientist at the Olklahoma State University, analysed more than 800 movies released between 1998 and 2002 and came up with a "neural recognition" programme that judges films based on seven parameters,
The criteria were: 'star value' of the cast, the movie's age rating, the time of its release against that of competitive movies, the film's genre, the extent of special effects, whether the film is a sequel or not and the number of theatres where it is to be premiered.
Analysing these through the programme, Sharda was able to place the movie in one of the nine categories ranging from 'flop' (with total takings less than $1 million) to 'blockbuster' (over $200 million), British weekly New Scientist reported.
The programme, when tested on past movies, turned out to be correct 75 per cent of the time to within one category either side of the correct answer. Out of these, 35 per cent were spot-on.
{{/usCountry}}The programme, when tested on past movies, turned out to be correct 75 per cent of the time to within one category either side of the correct answer. Out of these, 35 per cent were spot-on.
{{/usCountry}}Sharda, who graduated from Udaipur University, is now working on the software with a "major Hollywood studio," besides expanding the system to include DVD sales.
He is also building a website where users can read a movie's forecast by entering its key parameters.