The first documented public screening of a film (short film) was in 1895 and the distinction goes to the french brothers Auguste and Louis Lumière.
There was total excitement in Paris when that happened and about 200 people paid to watch 10 short films on 28th Dec 1895.
As this nascent art was taking shape, and more people began arriving at auditoriums, the first artistic critique of films were published in 1908. In its early forms it was a gentle criticism of films and was passionately consumed by the growing audience for this new art form.
Since the narrative style of criticism of films are long, Irene Thirer, the film critic of the New York Daily News, in the 31 July 1928 issue of the the newspaper began grading movies on a scale of zero to three stars. Three stars meant ‘excellent,’ two ‘good,’ and one star meant ‘mediocre.’ And no stars at all ‘means the picture’s right bad,’”.
A star rating and/or the narrative style of critical commentary on films were the opinion of one person, and so very subjective. So to solve for that, in 1950 the french film magazine Cahiers du cinéma “started polling critics and boiling their judgment down to a star rating. The highest rating any film earned was five stars. This helped eliminate concentration of critical powers with a few individuals, and movies that were universally liked or dis-liked tended to get noticed better with this approach.
As computers came along and digital transformation happened during the last few years of the late 20th century, imDB, Rotten Tomatoes and Meta Critic used this new power to aggregate film reviews both from critics and users automatically. Users had access to internet and these companies used the computing power to build sophisticated statistical models, and scored movies based on millions of user’s opinion. This helped universally popular movies to standout —the herd knows the best gracing grounds. Though a democratic approach helps people of all race, color and status have a voice from a political stand-point, it tends to mute any actionable movie insights. If two users scored a movie 0 and 100, the average score for the movie is 50. This makes the average score totally non-actionable if you happen to have a taste that is left or right of center.
As the time machine reached the machine learning era, CineBee re-imagined reviews once again to be individual centric and not herd centric by computing every user’s taste. CineBee understand’s the user’s taste from their reviews and using state-of-the-art graph and neural models, finds movies and TV shows that matches their taste. CineBee also provides a true 360 aperture view with continued support to critics reviews and also herd scores for movies and TV. CineBee also identifies which critics have the best taste chemistry with each user by ordering them appropriately.
CineBee also maps the user’s taste with tastes of other movie and TV aficionados and creates a personalized tribe for every user. Using some deep collaborative filtering techniques, CineBee harvests the Tribe collective intelligence to provide both prediction and recommendation to any user.
With the new streaming revolution where there are hundreds of streaming platforms with vast libraries, CineBee is built for every user to get precise recommendations and prediction across various OTT platforms using their taste and affinity as the index.
Conclusion: Movies reviews have continued to morph, from the silent movie era → commercial cinema era → computers era → cloud computing era with the one philosophy of providing accurate and unbiased recommendations. With the vast availability of computing power, now we are able to draw attention to the user’s taste to provide unbiased and accurate movie and TV recommendations.