In terms of the technology we need to optimize our servlet algorithm by providing simultaneous (using Java threads) requests to social network APIs eliminating in that way the latency. Further improvement is to provide a buffer for each request for example the last 20 movie searches can be cashed on the servlet so in case the same query-string is requested again then the will return the information directly without making the request to the third party. This happens frequently because the user type a name on the search bar and the presses the backspace to correct it. The string after pressing the backspace is the string had previously requested.

We consider vital the integration of you system with foursquare because our system is mostly targeted foursquare type of users. This kind of integration will have benefit all parties (user, Facebook and MovieIt) because the user gets better recommendation (from MovieIt) and badges (from foursquare) the Foursquare and MovieIt will get information from the user activity the former for places and the later to provide better recommendation.

Finally there is a need of automatic composition, integration, and execution of Web Services. We can achieve this by extending our system to use Semantic Web Services based on RESTfull Web Services. This Semantic web technology will facilitate integration of other systems without ours gaining in that way more information for other sources and therefore more value.

Growth rate ( Image taken from venturebeat.com )

Thanks to the astonishing growth of the internet, having millions of users is no longer a dream for new applications.  It used to take applications years to get more than a hundred thousand users, let alone millions. But nowadays it’s not very unusual for applications to attract millions of users in couple of months (or even in a few days).  So that, being prepared for this huge number of users and having an effective structure for dealing with this large numbers is essential for designing social applications. After all, nothing kills an application’s growth more than being down for a couple of days (or even a few hours).

Therefore we have designed our application with the scalability issue in mind.  Scalability issues that we have considered are in two categories of technical issues and development issues.

The main issue that we have investigated in the technical issue is the issue of database. We have decided that MySQL is an effective DBMS for starting the application and gaining the first few thousands users. However, we are aware that as our user base grow beyond those numbers, MySQL becomes less suitable. Therefore we have decided to move from Relation database management solutions to database systems which offer eventual consistency, at that time.

The development issue that we have considered is that when we attract  more users to our system, we would need more programmers to work on the application and develop it. Therefore we have used standard development techniques in order to minimise the effort the new programmer’s effort to get to know the system.

Interest Graph( From Interest Graph( Picture by ShinyShiny.tv)

Om Malik (senior writer for Forbes.com, Red Herring and Business 2.0), in his recent blog post, explains his views on the next generation of social networks.
He predicts that, in the future,  there will be lots of specific social graphs, instead of one big social graph.  He explains that  our friends are not necessarily the best source of recommendations in all aspects. We might have a friend that we really like his/her taste in music, but we don’t like his/her sense in movies.

Therefore he suggests that the next generation of social applications are those that build their interest-oriented networks on top of large social networks. Furthermore, he thinks that these kind of networks have more potential for making purchases happen (something which social networks advertising is not currently very good at, in compare with other online industries such as search engine advertising).

For entrepreneurs, I believe there are opportunities to create unique experiences around the concept of “interest graphs” that can be built off the backs of uber-networks such as Facebook and Twitter. These networks can help find the right kind of audience to build a viable channel for new commerce experience. (Om Malik)

Our movie application (MovieIt) has been built on the very same idea. Users in our application have the opportunity to choose the people they want to follow, and connect with the ones they trust, instead of following all their friends.  Furthermore,our application has a large potential for making purchases happen ( which in our application is buying cinema tickets).

Critics

1. No ways to receive profit (possibly project is not attractive for real world investors);
2. Security and ethical issues (How much information about user preferences and visited places it should expose? Possibly, some users would like to hide information about visited places and who was with them at this time.)  ! However, these drawbacks are intrinsic to all social networks.

Suggestions

1. In order to improve the quality of recommendations system should memorize and analyze user’s preferences (notice what friends you follow more often, how you replied on received notifications: liked it or disliked, how you rated suggested movies and cinemas, etc.).  Benefits of this “tastes history”:
   - more personalized recommendations;
   - than longer user exploits an application than more relevant suggestions it delivers (extension to trust-based algorithm).
2. System should support reversing scenario: user is entering the place (cinema, for example) and application provides suggestions what are the best options to watch in this place. (“Here I am. What should I see?”)
3. It is possible that you and your friend have completely different tastes in movies, so making a group of friends with similar tastes was suggested. (We can also have a list of friends and rank them)
4. System should stimulate best contributors (discounts in cinemas, for example).
5. System should provide an opportunity to invite friends on future events (tag your friend if you want to go and watch a movie together).
6. System should be highly responsive.
7. System should be extensible.
8. Developers should think about ways, features which might be interesting for potential investors (advertisement, online movie purchases, etc.)

As we discussed the system should not depend on Facebook or Twitter. Twitter is just a better way to get people and twit our friends.

Legal issues was another topic in our discussion. As we mentioned before the system allows its users to post their comments and reviews about cinemas. According to cinema policy, cinema can respond to user comments and asks to remove unfair reviews.

Geo locations, business part of the system and reliability were other concepts of our discussion.

Summary

Considering results of the discussion we can conclude that our “potential users”/reviewers were interested in proposed system and admitted that they would like to use such kind of suggestion tool. However, they pointed out that even though system has a social value it might be unnoticed by investors as it doesn’t provide good capabilities for receiving profit.

Finally, we would like to thank our “critics” for their system evaluation and suggestions. We have found this session very useful and got very good ideas about future concept evolvement. Moreover, we are going adjust our system functionality according to the received feedback and include some presented above suggestions in our application functionality list.

As mentioned before, our proposed system is a movie recommendation system that tells its users what movies to watch and also who has suggested that movie. It also suggests its users the best place for them to watch the movie using geo-location info. Another feature of our system is that it lets its users to know who from their friends might be interested in watching the movie.

In order to estimate the usability of our system, we conducted a survey to check other people`s opinions about our system. What the results show, encourage us to work harder on our system.

About 30 people took part in our survey. 94% of them are interested in watching movies. 42.31% of them prefer to watch a movie with other people at social places and 33.33% of them prefer cinema instead of home or any other places. 59.26% of them go to the movies with their friends. Internet is the best way for 59.26% of them to choose a movie to watch and 25.93% of them use their friend recommendations. More than 95% of them show interest in telling or recommending their favourite movies to their friends. Almost all of them have Facebook accounts and more than half of them want to use Facebook as a way to tell their friends about the movies they have watched and enjoyed. Some of them use their mobile phones to search for a movie or cinema. More than 80% believe that the location of the cinema is very important for them and they want to tell their friends about the location and quality of the cinema they have been to.  

To check the results click on the rest of this entry…

- Jinni: Jinni is a Taste Engine. It looks at film through the lens of what makes you love or hate anything you watch. With a Taste Engine, you don’t search by what you’re looking for, you search by what you like. And recommendations are based on analysing your preferences, not statistics. Registration is required for social networking aspects of sites.

- Tastekid: We input a film we like and it returns results that offer similar action to our favorites.

- Filmaster Blog: It is a free and open film recommendation API. The API enables external programmers to create independent services or apps using our data and algorithms, it allows to easily integrate any website with Filmaster by presenting our content like reviews and film recommendations.
Filmaster Mobile recommends you films to watch in local cinemas and TV. Wherever you are you can check-in to screenings and shows, meet film buffs with similar interests in your area and tell your friends what you’re watching. It’s local, mobile and based on personalized recommendations. In a nutshell: Foursquare for film.

- Filxup: Filxup is a movie recommendation app built for iPhone. Flixup claims itself to be the best source of what movies people are talking about on Twitter. The app measures people sentiment pertaining to a particular movie and decides if the movie is liked by twitter users.

- Faves for facebook: Favebook helps us to discover our friends’ favorite items and recommendations – with a few simple taps. Recommendations on You Tube, Wikipedia, Google, Amazon, and other websites can be also checked out. And best of all, Favebook tell us the most popular items in your Facebook friends’ profile.

- Movies Now: Movies now helps us find movies that are playing around us. We can also browse showtimes plying around us within next 2 hours in a list organized by the most convenient movie for you to get to. We can also purchase our movie tickets anywhere on the go.

Recommendation systems and algorithms

The growing need of a good suggestion tool in the web of data boosted the development of different sorts of recommendation algorithms. Most popular and widely used are collaborative filtering (CF) and item to item recommendations. The first one suggests items exploiting the similarity between users when the second one provides recommendations measuring the similarity between items. To some extent both algorithms work quite well and produce good quality suggestions, however, they suffer from the number of limitations as impossibility to include new items which were not rated in the final suggestion list or take into account recently registered users. In order to overcome these drawbacks authors of [1] research proposed to use standard recommendation algorithms in conjunction with social networks infrastructure. They considered what suggestion algorithm can benefit from being used within the social network and their answer is definitely – collaborative filtering as it can consume existing information of user profile. However, researches stress that the pure CF recommendation based on the measurement of resemblance among all system users is less trust-worthy than the suggestion made by friend or family member. In the other words, paper proposes to utilize classical CF approach, physical (user location, time and environment) and social (interests, friend’s lists, etc.) contexts of user profiles in order to provide more precise and trust-worthy suggestions to users of mobile devices. Research team developed live demo and ran a set of tests which confirmed high effectiveness of proposed solution.

Some ideas expressed in a previous research were reflected by Chinese researcher Lixin Zhou [2] who also considered trust as a powerful tool for enhancing the effectiveness of any recommendation system. In order to provide precise recommendations he proposed to exploit Social Network Analysis technique which allows to extract valuable information from connections between users and recommend items with a reference to a friends list. In support to his proposal a prototype of trust-based recommendation system was developed. However, research paper doesn’t provide any empirical data which could prove the effectiveness of a new method in comparison with a standard CF.

Next research [3]  utilizes the Facebook Social Network infrastructure and API’s provided by third party web sites like www.maxmind.com for geo location information, www.last.fm for events gathering in order to make event recommendation based on the content that events (Content Based Method) have and the rating that users assign to them (Collaborative Method). Based on this information the system is able to make recommendation to a Facebook user and also suggest friend that might be interested in attending the selected event.

In addition to described above works [4] and [5] also consider benefits from using CF in conjunction with social networks infrastructure and trust weight and provide evidences of its successful use.

Recommendation quality factors: contribution and participation

Another interesting aspect which affects the quality of recommendations is the willing of social community to contribute and participate. “Social comparison and contributions to online communities (the MovieLens study case)” paper [6] raises questions about “under-contribution” and “non-participation” of social network members. Current study revealed that in content-sharing systems, for example, 25% of users share 98% of the content while the rest (75%) only consumes. This state-of-art creates a serious problem for recommendation systems as they rely heavily on user’s ratings and reviews in order to provide high quality recommendations. In follows researches questioned what can motivate community members to contribute? Results of a study showed that after notifying users about their social network behavior and the level of participation “passive” users show higher level of contribution (530% increase) while “active” users slow down their activity (62% decrease).

Presentation logic

The last paper in the current survey set considers the effect of a suggested items presentation on user’s persuasion and satisfaction. In the other words, researches focused on presentation logic of recommended items: they tested different presentation methods as N-items lists,  structured overview and recommendation modality (i.e. simple text, combination of text and image, and combination of text and video). Empirical evaluation of different methods revealed that the most efficient presentation modes are structured overview and recommendation modality (text + video or text + image).

Structured overview

Text plus image and text plus video presentation modes

Summary

Considering collected information we can conclude:

1. The use of a social networks infrastructure (relations between users, profile information and location data) in coherence with standard recommender algorithms allows to provide more qualitative and trust-worthy suggestions;
2. In order to support recommendation systems we need a high level of participation (reports about social activity and comparison with other users are needed);
3. Presentation logic plays an important role in user’s persuasion and satisfaction;

Features which might be useful in our project

1. The use of social networks infrastructure for recommendations;
2. Reports about social activity and the level of contribution;
3. The use of structured overviews or modality methods for recommended items presentation.

What’s next?

In this post we considered works related to recommender algorithms on top of social networks and factors which have an influence on the quality of suggestions as presentation methods and the level of participation. In the next post devoted to literature survey we will cover topics as geo location systems in social networks, the level of anonymity and privacy issues, legal issues with user-generated content and social and economic impact of recommendation systems.

References

1. W. Woerndl, G. Groh, “Utilizing Physical and Social Context to Improve Recommender Systems”, IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology, 2007.
2. L. Zhou, “Trust Based Recommendation System with Social Network Analysis”,  Information Engineering and Computer Science International Conference, 2009.
3. Kayaalp, M. Ozyer, T. Ozyer, S.T., “A Collaborative and Content Based Event Recommendation System Integrated with Data Collection Scrapers and Services at a Social Networking Site”, International Conference on Advances in Social Network Analysis and Mining, 2009.
4. Zhang, Ruiquan Sun, Baohui Kang, Wei Zhu, Tingshao, “GLOSS: A Social Networks-Based Recommender System”, 5th International Conference on Pervasive Computing and Applications (ICPCA), 2010.
5. Faqing Wu; Liang He; Weiwei Xia; Lei Ren, “A collaborative filtering algorithm based on Users’ Partial Similarity”, 10th International Conference on Control, Automation, Robotics and Vision, 2008.
6. Y.F. Chen, M. Harper, J. Konstan, S.X. Li, “Social Comparisons and Contributions to Online Communities:
   A Field Experiment on MovieLens”, Ph.D. Thesis, University of Minnesota, 2009.
7. T. Nanou, G. Lekakos, K. Fouskas, “The effects of recommendations’ presentation on persuasion and satisfaction in a movie recommender system”, vol.16 Multimedia Intelligent Services and Technologies,  2010.