A method for scheduling the display of content

Abstract

A method for scheduling a plurality of films or movies at one or more cinemas, comprises; creating an individual taste profile for customers or users, the taste profile including distance to each cin

Description

Field of the invention The present disclosure relates to a method for scheduling the display of content to paying users, for example scheduling the display of a plurality of films at a plurality of cinemas for a plurality of users. Background Large groups of people often enjoy a form of pre-recorded visual or auditory entertainment together. This entertainment can be films, TV shows, music, live sport events, etc. Others include visual arts in Cinemas, drive-by cinemas and theatres. This also applies to upcoming music in a bar or pub or anywhere where music is playing for a group of people. The scheduling for these events is typically fixed in advance, and the viewer gets little or no input. This disconnect leads to a lack of engagement by said people, and lowering their overall experience. Fig. 1 is an illustrative representation of audience-controlled entertainment scheduling, showing the traditional technologies available before the introduction of the present disclosure, as a system diagram. The methods for watching TV 107 or listening to radio 101 are presented as system diagrams. In summary, both methods provide the audience with different modes of listening and watching what they want and when they want - in line with how much they are willing to pay. For example, an audience member who wishes to watch a certain TV program 108 will not chose to watch regular TV 111, but instead may choose to pay extra for it 109. Therefore, if they want it on demand 110, they may choose to watch it on demand 112 via online streaming services, or may choose to use slower means such as ordering a DVD 113. The present disclosure aims to improve these processes and apply them to cinema viewings as well as live music. Fig. 2 is an illustrative representation of various audience controlled entertainment scheduling systems, showing how radio 201, TV 202, and now with the present disclosure, movie theatres 203 have been given dynamic scheduling, as a system diagram. The different mediums of entertainment are shown, as well as the evolution they have followed into dynamic viewing and scheduling, such as music streaming services 205, film streaming services 206, and now a new system for movie theatre screening 207, all providing the power of choosing what is played 204. Summary of the invention Aspects of the invention are as set out in the independent claims and optional features are set out in the dependent claims. Aspects of the invention may be provided in conjunction with each other and features of one aspect may be applied to other aspects. In one aspect, there is described a method for scheduling a plurality of films at a plurality of cinemas for a plurality of users, the method comprising the steps of: creating a taste profile for each user, the taste profile comprising proximity to each cinema and film preference information; determining a likelihood of success by a processing system for each film at each cinema based at least in part on the taste profiles for each user; and generating a viewing schedule for each film at each cinema based at least in part on the likelihood of success for each film. Advantageously this allows the marketing of films and creating a viewing schedule for films before any major distribution deal has been agreed, meaning a better deal can be obtained by all parties. Optionally wherein each cinema is in a different location. Advantageously, this means that each cinema will have a different proximity to each user. Optionally wherein determining the likelihood of success is further based on a cinema success history for each cinema, wherein the cinema success history comprises ticket sales for previously screened films and film information for previously screened film, the film information comprising a film genre and a film description. Advantageously, this accounts for historical film success at each cinema. Optionally wherein each cinema has a plurality of screens having a plurality of slots for screening films, and the viewing schedule populates each slot of each screen at each cinema with one of the plurality of films. Advantageously, this allows scheduling of films in multiple time slots, possibly accounting for a full days programming or more. Optionally the method further comprising the step of: testing demand for each film and altering the viewing schedule based on the test. Advantageously, this means the demand for each film can be tested before making a final viewing schedule, helping to more accurately meet demand. Optionally the method further comprising the steps: generating a distribution deal for each film based at least in part on the viewing schedule; generating deals for each user based at least in part on the viewing schedule, the distribution deal, and the user taste profile. Advantageously, this means that the method handles such distribution deals, rather than people physically negotiating, therefore saving money. Optionally wherein generating a distribution deal for each film is further based on a distributor contribution. Advantageously, a distributor may be inclined to pay more or less to have a film shown to more users or less users respectively. Optionally wherein generating deals for each user is further based on the likelihood of each user watching each film at each cinema. Advantageously, this means that deals can be generated for each user depending on whether they are deemed likely to watch each film. For example, if a user is likely to watch a film, they may be given a better deal to watch the film, or conversely, if a user is not likely to watch a film, they may be given a better deal to help convince them to watch the film. Optionally wherein the likelihood of success for each film at each cinema includes a likelihood of each user watching each film at each cinema. Advantageously, this accounts for whether a user will actually view a film at a certain cinema. For example, a film will have a lower likelihood of success at a cinema further from a user than a cinema closer to the same user. Optionally wherein the user taste profile further comprises at least one of a preferred screening time, a preferred cinema location, a film viewing history, a film preference history, a user follow through rate, and a film cost preference. Advantageously, this can provide an in-depth profiling of what a users taste in films and viewing times is, enabling a better match for films to each user. Optionally wherein the method also includes the step of: surveying at least one user of the plurality of users on at least one of the films, and wherein determining the likelihood of success is further based upon the survey. Advantageously, this means that marketing of films and discovering which films seem to be popular occurs from the very beginning, before a major distribution deal may be agreed. Optionally wherein the survey checks at least one of whether the user has already seen the film, whether the user would prefer to watch the film in a cinema or at home, whether the user wants to see the film, whether the user is willing to spend more or less than a nominal cost. This allows the survey to gather further information about the user, better informing the likelihood of success for each film. Optionally wherein for each film the user has not seen, the survey presents a trailer for that film being surveyed before checking whether the user wants to see the film. This provides the user with information about a film before asking them to provide an opinion on whether they would want to see the film. Optionally wherein the viewing schedule may further include a voting system for scheduling films during a voting slot at at least one cinema to be voted on by ticket holders. Advantageously, this allows a crowd of people, both users and non-users, collectively known as ticket holders, to select the film that they want to see. This may be further facilitated by non-users by devices provided in the cinema. Optionally wherein the voting system comprises at least two stages, the first stage providing information on a selection pool of at least three films to the ticket holders who then vote on at least one film of the at least three films; in the second stage, the least popular of the at least three films is removed from the selection pool, and the ticket holders then vote on at least one film of the now at least two films, with the remaining most popular film being chosen as to be viewed in the voting slot; and in the event of a tied stage where there are at least two films with the least number of votes, the tied stage is re-run with each ticket holder able to vote on one less film than they were previously able, unless that would mean each ticket holder would not be able to vote at all, in which case the selection pool will be altered and the voting system restarted from the first stage. Advantageously, this voting system progressively eliminates films from the selection pool until only one remains. It may also be the case that there is only 1 stage of voting between two films, and in the case described, other selection or voting methods may be used to select the film to be viewed. Such a method may be just one round of voting where the most popular film gets viewed, or in the case of a tie, there is a further vote between only the tied films. In the event of a further tie, a ticket holder may be randomly selected to not be eligible to vote (i.e. if there is an even number of ticket holders). Optionally wherein each stage may be open for voting for a selected period of time, and / or may last until at least a minimum number of ticket holders have voted. This allows different methods to end a voting stage, such as by timing out the voting, or if enough ticket holders have voted. Optionally wherein at least one ticket holder is a user. Advantageously, this means that if there are no devices available to use for voting at a cinema, at least one user is required to be a ticket holder in order to vote on a film. Optionally wherein the processing system is based at least in part on a machine learning model. Advantageously, this means the system may continuously learn and improve the accuracy of determining the likelihood of success as well as all other areas. In another aspect, there is provided a method of training a machine learning model for use as a film success indicator for at least one film at a plurality of cinemas for a plurality of users, the method comprising the steps of: receiving film information, the film information comprising a film genre, a film success metric and an average user score; receiving a plurality of user taste profiles, the user taste profiles comprising proximity to each cinema and film preference information; and determining a film success indicator for each film at each cinema based at least in part on the film information and the plurality of user taste profiles. Optionally wherein the method further comprises the step of surveying at least one user on at least one film that best matches their user taste profile, and wherein determining the film success indicator for each film is further based on the survey. In another aspect, there is provided a method for scheduling a service at a plurality of service providers for a plurality of users, the method comprising the steps of: creating a user profile for each user, wherein the user profile comprises user location information and service preference information; determining a likelihood of success by a processing system for each service at each service provider based at least in part on the user profiles for each user; and generating a service schedule for each service at each service provider based at least in part on the likelihood of success for each service. Drawings Embodiments of the disclosure will now be described, by way of example only, with reference to the accompanying drawings, in which: Fig. 1 is a representation of audience-controlled entertainment scheduling, showing existing technologies. Fig. 2 is a representation of audience-controlled entertainment scheduling, further showing existing technologies, and how the present disclosure compliments these. Fig. 3 is a representation of audience-controlled entertainment scheduling, showing an exemplary algorithm of the present disclosure for generating a viewing schedule. Fig. 4 is a representation of audience-controlled entertainment scheduling, showing an exemplary voting system for voting on films in voting slots. Fig. 5 is a representation of audience-controlled entertainment scheduling, showing an exemplary survey system using an application on a mobile device. Fig. 6 is a representation of audience-controlled entertainment scheduling, showing how multiple cinemas may have multiple screening times, including voting slots. Fig. 7 is a representation of audience-controlled entertainment scheduling, showing how the present disclosure may be applied to various other applications. Fig. 8 is a representation of an example audience controlled entertainment scheduling system, showing how various of the other systems described herein may be combined. Specific description Fig. 3 is an illustrative representation of audience-controlled entertainment scheduling, showing exemplary factors used in the algorithm and outputs in a viewing schedule, as a system diagram. The system diagram 300 shows the process of acquiring user data for the scheduling algorithm 300. As shown in fig. 3, there is an exemplary method for scheduling films at cinemas for an audience, who in other words may be users of a system or app. Each user has an associated taste profile that has various pieces of information about the user and their preferences. For example, the taste profile contains at least the users proximity to each cinema, optionally in the form of a home location or a set location 304. This location can also be dynamically via a device of the user such as a mobile phone. The taste profile also contains at least some film preferences 301 of the user. This information can contain information regarding films the user has previously watched, as well as an associated score of whether the user enjoyed the film or not. Other optional pieces of information can include the users time and day preferences 302 for watching a film, i.e. on the weekend in the morning, or weekdays in the evening, the users seating preferences, and other behaviors and tendencies 305. There may also be a suggestion from the user as to what film to show 303, and a list of films the user wants to see 306. Feeding the taste profiles into a processing system or algorithm 307, along with the relevant cinemas and films to be shown or screened, the algorithm can determine a likelihood of success of each film at each cinema. Subsequently, this likelihood of success can be used to generate suggestions of what films should be screened 308 and a viewing schedule 309. This viewing schedule can include for each film location suggestions 312 (i.e. which cinema to be shown at), time suggestions 311 (i.e. what time of day the film is shown at), specific screens at a cinema 314 (e.g. if there are screens with larger or smaller capacity) as well as other features such as price suggestions 313. The algorithm can also use information from each cinema as to the success of previous films at said cinema. This may be particularly useful, as not everyone who sees each film at a cinema will be a user on the app. Therefore, while active users taste profiles will be used, the historical performance of films at a cinema will still account for viewers who do not use the app. The historical performance of films can include ticket sales for previously screened films, as well as information regarding the films themselves, such as the film genre, title, description, age rating, percentage of viewers who were also users, etc. An example of how this and other features may fit together is shown in fig. 8. Once a viewing schedule has been generated, a distribution deal for each film can be generated and proposed based on that viewing schedule. For example, if a film has been scheduled to be shown more, the distribution deal may propose paying a higher licensing fee to the distributor. Further, it is also possible that a distributor may even pay a fee to have their film either surveyed more by users or to even have their film simply screened more. In the latter case, a distributor paying a fee will increase their films likelihood of success and therefore increase how much the film is screened in the viewing schedule. User deals may also be generated based on a combination of these factors, such as the viewing schedule, the distribution deal, and the user taste profiles. For example, if a distributor pays additional fees, their film may also be advertised to users at some discount relative to the additional fees paid. Additionally or alternatively, if a film closely matches a user taste profile, that user may be offered a unique deal to buy tickets to see that film. Further, the likelihood of success metric may also include a likelihood of each user watching each film at each cinema, which may then further influence the viewing schedule. For example, if a film closely matches a user taste profile, then the likelihood of success may be quite higher for this film to this user, but if there are no cinemas close to the user, then this films likelihood of success is not significantly improved. Conversely, if a film somewhat matches every users taste profiles, and each user has cinemas very close to them, then the likelihood of success for that film may be higher than the previous scenario. Further, a survey or active voting system may also be used to select what films to show in a cinema. With reference to Fig. 4 is an illustrative representation of audience-controlled entertainment scheduling system, showing the process of an active voting system, as a system diagram. The system diagram shows a typical voting system as followed by the user. In order for the voting system to be effective, at least one screen at a cinema must have at least one available slot to be used as a voting slot. In other words, the film to be shown in this voting slot will be decided by the voting system. Users, or general cinema visitors, can purchase refundable or exchangeable ticket 401 to see a film in this voting slot, with the knowledge that they will be seeing a crowd-voted film. They are then shown film information for a selection of films, such as the title, a brief description, and optionally at least one trailer. Each ticket holder can then vote 402 on whether they would like to see each film. At this first stage and in subsequent stages, that means a ticket holder is able to vote that they would be happy to see all of the films. This first stage will last a nominal amount of time, such as one hour, or if a minimum number of tickets have been sold, then until each ticket holder has voted. Subsequent voting stages will then begin, cutting out the least popular film from the previous stage 404. If there happens to be a draw between all film voted on in a stage, i.e. if there are 3 films to be voted on in a stage and they all attract the same number of votes, then the stage is repeated, with the limitation that one less film may be voted for. In the example of three films attracting the same number of votes in a stage, then the stage will be re-run with each ticket holder only being able to vote on two films. In the final stage between the two remaining films, whichever film is selected will be shown during the voting slot 405. At each voting stage, each ticket holder must re-vote 403 for each film. Each stage may last a different or same nominal amount of time, such as each stage lasting an hour, each subsequent stage lasting less time than the previous stage. In some cases, each stage may end once a minimum number of ticket holders have voted, or if there are a minimum number of ticket holders such as 50, then once each ticket holder has voted the stage may end. Each stage may use any of these closure mechanisms, for example, all stages other than the final stage may end upon reaching a nominal time, but the final stage ends once each ticket holder has voted. If a ticket holder is not a user, they may be provided with a voting device at the cinema. In some examples, the ticket is refundable or exchangeable, meaning that if at some stage during the voting process a ticket holder does not see any film that they would vote on, they may instead choose to refund their or exchange their ticket 406 to a ticket of equivalent value, or to a ticket of different value where the difference is either provided or refunded to the ticket holder depending on the new ticket value. For example, if a ticket holder no longer sees their film of choice in a voting stage, they may choose to exchange their ticket to another voting slot ticket and begin the process again. Additionally, or alternatively to the voting system, there may also be a surveying system, such as that described in Fig. 5 as an illustrative representation of audience controlled entertainment scheduling, showing the interface of an embodiment of the invention (Ie prince), as a system diagram. The system diagram represents the interface the user will see if they use the app, which may include the voting system as described in fig. 4. In this part of the platform the audience can give their suggestions 503 and opinions 513 on films and showtimes. In some examples, this is to be inferred by giving a thumbs up or pass to a film clip or image. The user can be shown 501 504 film clips, trailers, images, descriptions, and other film information regarding various films. Once the user opinion of the film is given, further film information can be shown about a different film or same film. With this information collected, it can be combined with the voting system to provide an insight into what films are popular and inform which films should be used for the voting slots to be voted on, and can be tailored based on what users buy a voting slot ticket. For example, if a typical film roster for a voting slot is film A, film B, film C and film D, but a user buys a voting slot ticket with a desire to watch 514 films E and F, then the film roster for the voting slot may further include film E and film F, or the two additional films may replace two of films A-D. This replacement may happen by removing the least selected film in a nominal amount of past voting slots. For example, if films B and D have been the least popular films in the past 5 voting slots, then films E and F will replace these. Further, the survey information may be further used to inform the algorithm 517 as mentioned above. Each user may be able to change their preferences 510, such as their preferences in what time they go to a cinema 511, what cinemas they would go to 512, and any list of films that they want to see 513. All of these may be included in a user taste profile, or other user information that can be used by the algorithm 517 to produce a cinemas film schedule 518. Further This can also be used to provide personalized deals for each user 519. The voting system and survey system may also be combined to make a hybrid voting system, whereby the voting is performed by the algorithm using the user preferences created and updated through the user interacting with the app, building a more complete view of what the user's preferences are. Then, in cases where all ticket holders are users, the voting system may be run as described above, but 'voting' by matching each users user preferences to their perceived 'favorite' movie in the movie roster, and narrowing down the film to a single film. If no film is particularly matched to enough of the ticket holders user preferences, then the ticket may be cancelled. Additionally, in any of these examples and cases, a user may also be able to invite other users to buy a ticket to their screening. Further, in the hybrid voting system, once a film has been chosen, all ticket holders may then be notified that a selected film is being screened. Fig. 6 is an illustrative representation of audience-controlled entertainment scheduling, showing the cinema schedule for an example cinema 637, as a system diagram. The system diagram shows a typical cinema schedule in line with the benefits of examples and embodiments described above. For example, at the typical cinema with 5 screens 601-605 and three times 606-608, there are still fixed slots 615-621 where a film is pre-decided, but there are also voting slots 622-627 where a film is voted 635 upon as described above. In the case of fixed slots, these can also be decided by the algorithm as discussed above, and further informed by the surveys. For example, if there are 8 films from voting 633 and additional films suggested from user preferences 634 being considered by the algorithm, and they are ranked according to user taste profiles, then a predicted success of these films can be obtained. However, the survey may refine the predicted success and taste profiles, and demand can be tested through the surveys. Both approaches may be used separately or in combination. For example, the films 633 and 634 can have a predicted success by the algorithm, which then surveys a plurality of users, where at least a portion of this plurality had their user taste profiles considered when predicting success. The results of this survey can then be used to re-predict success. This would be testing demand. An example of using the survey to refine the predicted success without testing demand would be to survey a plurality of users on the 10 films, and then user a portion of the plurality of users taste profiles to predict success of each film. It would also be possible to combine these approaches, refining the predicted success, testing demand, and further refining, by using a plurality of different film information for each film at each stage of this combined approach. Overall, it would still be possible to book a confirmed ticket 631 to a specific film , but also a possibility to book a voting ticket 632 to vote 635 on what films 633 and 634 each user wants to see. Further, testing demand may also involve surveying further users more on certain films or film information based on various factors, such as whether the film has been seen by a nominal number of users, whether a group of users like a film - so the film is shown to further users to confirm whether that film is popular within the wider population. Testing demand may also be implemented at the request of a distributor or other party for specific films. Fig. 7 is an illustrative representation of audience-controlled entertainment scheduling, showing how the present disclosure can be applied to multiple industries, as a system diagram. The system diagram shows the many industries the disclosure can be applied to, including live music 704, different kinds cinemas and physical film screening establishments 705, 706, 707, 708, 709, 711, and more. The disclosure may be applied to films, TV, radio, live music, or any other variations known in the art. Some parts of the disclosure may be in the form of an application (app), a desktop software, a cloud based software, or any other variations known in the art. The use of the present disclosure may vary depending on whether the use will be on-demand control by an audience in real time 703, or control by an audience but not in real time 702 such as when audience preferences are used, but to create a schedule in the future, rather than deciding what will be immediately viewed. In real time, instant group voting then screening the media 712 may be used, but when scheduling is not decided in real time entertainment selection by an audience can be voted on in previous days 713. Referring to Fig. 8, an illustrative overview of the system can be seen. Each user may have a taste profile 808 compiled by an algorithm 807, optionally by K-means clustering, collaborative filtering, hybrid filtering or other machine learning algorithm, where the algorithm 807 has access to each users film preferences 801 (likes or dislikes), their film viewing time preference 802, their location 803, their activity and behavior 804, their cinema going history 805 and their connections to other users 806. Also fed into the taste profile 808 for each user is the interest gathered towards a film 825 for a film recently added to the app / platform 824. If there is enough interest 827 from each user taste profile in a film added to the platform 824, after some delay 826, there can also be a check if that film will be successful if it is screened 828. If not, the data will be stored with information on which niche audience is interested potentially for a later screening 829. Otherwise, the new film 824 will be shown to more users to find the best time, location, price, users, before suggesting which cinema it should be shown at 830. At 843, a film starts being shown to users, gathering information regarding whether they would like to see that film. Such examples of these films would be films such as those covered by 830 into 842 as a newly added film that is becoming popular, a newly released film 838, a user suggested film 839, a film suggested by a filmmaker who wants to test demand for a film 840, a film suggested by a cinema to test interest for the film 841. From this information 843, the film can be shown to users to find hotspots and niche groups that would like the film 844, and find users who want to watch the film and categorize them in their likelihood of attendance 845. Then, based on each users likely attendance for each user with this film, model the estimated attendance at each cinema at any time 846. Modelling estimated attendance 846 also uses the relationship 813 between each user, film and cinema. The relationships 813 are generated through algorithm 812 which in turn takes an input of which films each user might watch in the cinema 809, which cinema each user might visit 810 and how much eas user might pay for a certain film 811, which are taken from user taste profiles 808. Further, the algorithm 812 uses film information added to the app / platform 831, as well as input from the cinema(s) 823 including information such as each cinema's location, price range, type, opening hours and type of films they show 832. The distributors terms and conditions for exhibition of films, their preferred profit share and acceptable terms 834, along with the cinema's terms and conditions for exhibition of films, their preferred profit share and acceptable terms, as well as a specific film chosen for a screening 817 are then used to automatically generate a deal for exhibition 836. The specific film being chosen by considering a combination of the cinema's suggested films 814, likelihood of success of a screening the film at any time 815 the suggestion 816 of what films should be screened. Where the suggestion 816 of what films should be screened is obtained from the likely attendance each user has with this film, model estimated attendance at each cinema at any time 846 being processed to find successful screening opportunities 847. In the case films that do not have successful screening opportunities found, they will be logged as small unmet demands for users of a location and alternative screening opportunities will be found and displayed to other cinemas 848. Further, for each successful screening opportunity found 847, this will be used in combination with a database 852 of paths to successful screening for all other films that will be shown in the cinema or community, a cinema's pre-set rules for scheduling, making and generating deals 850, to model every screening possibility to find the best schedule 851. Additionally, with each successful screening opportunity found 847, this will be used for a specific time and place to calculate the likelihood of attendance for each patron of that cinema and categorise them in sections who would be likely to respond to a specific deal type 849. This in turn will be used in combination with the cinema's pre-set rules 850 and finding the audience members most likely to be interested in the film, when and where they are most likely to watch the film 835, to generate user specific deals and offers for cheaper tickets, last minute offers and more that could be set to automatically activate and target users in different categories (based on their attitude towards the film, cinema and the time) 837. Further, in finding the audience members most likely to be interested in the film, when and where they are most likely to watch the film 835, each of the perfect screen 819, perfect location 819, perfect price 820, and perfect time 821 can be selected. Further, the present disclosure could apply to various other live events that have some form of scheduling. For example, scheduling flights and deals for flights could utilize the methods taught in this disclosure, enabling the use of user taste profiles, surveys, and voting to help schedule flights more accurately to user demand, as well as a voting ticket to a selection of destinations, where the voting ticket may be cheaper than standard flights, or wherein the voting ticket may be a package ticket including other features and amenities such as accommodation in any of the voted destinations. Another example would be a changing and adapting the menu's in a restaurant based on user taste profiles and surveys as described above, as well as a voting system to decide certain items of the menu. Another example may also include creating a music or video playlist for large groups of people also using the above methods. It will be appreciated from the discussion above that the embodiments shown in the figures are merely exemplary, and include features which may be generalised, removed or replaced as described herein and as set out in the claims. Further, in the context of the present disclosure other examples and variations of the apparatus and methods described herein will be apparent to a person of skill in the art. Glossary of reference numerals for figures 5, 7 and 8 501- example application interface 502- comment functionality 503- suggest a film to the community 504- another example application interface 505- ability to show interest in a film 506- user icon, who's preferences may be updated 507- ability to double tap to express interest in the film 508- big database of film information and actors, directions, storyline and match score with user 509- film information 510- settings where a user can change their preferences 511- cinema going time preference 512- users "my cinemas" list 513- change list of films a user wants to see 514- list of films the user wants to watch in the cinema 515- rules and limits of screenings from the cinemas 516- conditions of filmmakers and distributors about their films 517- the algorithm 518- the cinema's film schedule 519- personalized deal for each user 520- example application interface showing personalized deals 701- control by audience 702- control by audience, but not in real time 703- control by audience, in real time 704- music listened to and enjoyed by a group 705- private or boutique cinemas implementing instant voting 706- cinemas 707- drive in cinemas 708- private cinemas used by a group (resorts or residents complexes) 709- cinema clubs, associations and societies (universities, schools, etc.) 710- theatres 711- pubs and events showing sporting events 712- instant group voting, then screening entertainment 713- entertainment selection by audience voting on previous days 714- entertainment selection by audience behavior and preference tracking algorithm and suggesting the most suitable option 715- choosing productions and events to be scheduled using audience preferences 716- entertainment selection by audience voting 801- User film preference (likes or dislikes) 802- User Time Preference 803- User Location 804- User activity and behaviour 805- User's cinema going history 806- Connections to other users 807- Algorithm (optionally K-means clustering, collaborative filtering, hybrid filtering or machine learning algorithm) 808- Taste profile for each user 809- Which films each user might watch in the cinema 810- Which cinema each user might visit 811- How much each user might pay for a certain film 813- Relationship between each user, film and cinema 814- Cinema's suggested films (the films that the cinema wants to screen) 815- Likelihood of success of a screening of any film in any cinema at any time is calculated 816- Suggestion of what films should be screened 817- Film chosen for the screening 825- Film starts being shown to users, Information gathered about interest towards the film 827- Where is there interest for this specific film and which users are interested 828- has there been enough users suggesting this film or indicating that they want to watch it that successful screening is likely? 829- Find niche audiences who are interested and store the data 830- Show to more users to find best time, users, location, price, suggest to relevant cinemas 831 - Film information added to platform 832- Each Cinema's location, price range, type, opening hours and type of films they show 833- Cinema's terms and conditions for exhibition of films, their preferred profit share and acceptable terms 834- Distributor's terms and conditions for exhibition of films, their preferred profit share and acceptable terms 835 - Find the audience members most likely to be interested in this film, determine when they are most likely to watch this film and which locations they might go to. 836- Deal for exhibition automatically generated 837- Generate user specific deals and offers for cheaper tickets, last minute offers and more that could be set to automatically activate and target users in different categories (based on their attitude towards the film, cinema and the time) 838- A film is newly released 839- Users suggests a film to be screened 840- filmmaker starts the process of further testing for a film 841- A cinema wants to screen a film and wants to find if there is interest 842- A newly added film is becoming popular 843- Film card starts being shown to users (Users are asked if they would see this film in the cinema?) 844- Find hotspots and niche groups would like this film 845- Find the users who want to watch this film and categorise them in their likelihood of attendance 846- based on the likely attendance each user has with this film, model estimated attendance at each cinema at any time 847- Successful screening opportunities found 848- Log the small unmet demand for the users of that location and find alternative screening opportunities for them and display this to other cinemas 849- for this time and place, calculate the likelihood of attendance for each patron of that cinema and categorise them into sections who would be likely to respond to a specific deal type 850- Cinema's pre-set rules for schedule making and generating deals 851- based on the predicted success likelihood of every other film for that time and place, model every 5 screening possibility to find the best schedule 852- Database of paths to successful screening for all other films that will be shown in the cinema or community 853- The schedule that services most patrons, allows most films to be shown or brings in most profits, or any combination of these

Claims

1. A method for scheduling a plurality of films at one or more cinemas for a plurality of users, the method comprising the steps of:creating a taste profile for each user, the taste profile comprising proximity to each cinema and film preference information;determining a likelihood of success by a processing system for each film at each cinema based at least in part on the taste profiles for each user; andgenerating a viewing schedule for each film at each cinema based at least in part on the likelihood of success for each film.

2. The method of claim 1, wherein each cinema is in a different location.

3. The method of any of the previous claims, wherein determining the likelihood of success is furtherbased on a cinema success history for each cinema, wherein the cinema success history comprises ticket sales for previously screened films and film information for previously screened film, the film information comprising a film genre and a film description.

4. The method of any of the previous claims, wherein each cinema has a plurality of screens having a plurality of slots for screening films, and the viewing schedule populates each slot of each screen at each cinema with one of the plurality of films.

5. The method of any of the previous claims, the method further comprising the step of:testing demand for each film and altering the viewing schedule based on the test.

6. The method of any of the previous claims, the method further comprising the steps:generating a distribution deal for each film based at least in part on the viewing schedule;generating deals for each user based at least in part on the viewing schedule, the distribution deal, and the user taste profile.

7. The method of claim 6, wherein generating a distribution deal for each film is further based on a distributor contribution.

8. The method of claim 6, wherein generating deals for each user is further based on the likelihood of each user watching each film at each cinema.

9. The method of any of the previous claims, wherein the likelihood of success for each film at each cinema includes a likelihood of each user watching each film at each cinema.

10. The method of any of the previous claims, wherein the user taste profile further comprises at least one of a preferred screening time, a preferred cinema location, a film viewing history, a film preference history, a user follow through rate, and a film cost preference.

11. The method of any of the previous claims, wherein the method also includes the step of:surveying at least one user of the plurality of users on at least one of the films, and wherein determining the likelihood of success is further based upon the survey.

12. The method of claim 11, wherein the survey checks at least one of whether the user has already seen the film, whether the user would prefer to watch the film in a cinema or at home, whether the user wants to see the film, whether the user is willing to spend more or less than a nominal cost.

13. The method of claims 11 or 12, wherein for each film the user has not seen, the survey presents a trailer for that film being surveyed before checking whether the user wants to see the film.

14. The method of any of the previous claims, wherein the viewing schedule may further include a voting system for scheduling films during a voting slot at at least one cinema to be voted on by ticket holders.

15. The method of claim 14, wherein the voting system comprises at least two stages, the first stage providing information on a selection pool of at least three films to the ticket holders who then vote on at least one film of the at least three films;in the second stage, the least popular of the at least three films is removed from the selection pool, and the ticket holders then vote on at least one film of the now at least two films, with the remaining most popular film being chosen as to be viewed in the voting slot; andin the event of a tied stage where there are at least two films with the least number of votes, the tied stage is re-run with each ticket holder able to vote on one less film than they were previously able, unless that would mean each ticket holder would not be able to vote at all, in which case the selection pool will be altered and the voting system restarted from the first stage.

16. The method of claim 15, wherein each stage may be open for voting for a selected period of time, and / or may last until at least a minimum number of ticket holders have voted.

17. The method of any of claims 14-16, wherein at least one ticket holder is a user.

18. The method of any of the previous claims, wherein the processing system is based at least in parton a machine learning model.

19. A method of training a machine learning model for use as a film success indicator for at least one film at a plurality of cinemas for a plurality of users, the method comprising the steps of:receiving film information, the film information comprising a film genre, a film success metric and an average user score;receiving a plurality of user taste profiles, the user taste profiles comprising proximity to each cinema and film preference information; and5 determining a film success indicator for each film at each cinema based at least in part on the filminformation and the plurality of user taste profiles.

20. The method of claim 19, wherein the method further comprises the step of:surveying at least one user on at least one film that best matches their user taste profile, and 10 wherein determining the film success indicator for each film is further based on the survey.

Images