| Date | 2nd April 2008 |
| Meeting Started | 8:55AM |
| Meeting Ended | 9:40AM |
| Members Present | Tabris Chen, Tristan Lawrence, Sandru Santhana Anandaraj |
| Week Number | 4 |
Team Four - Meeting Minutes
Business
- Project plan
The first goal for the team was to create a project plan as it was an assessable task. The following topics were discussed:
- Approach to the project
Tristan suggested that a mathematical approach be taken to analysing the data via Python. Essentially, by analysing trend lines on graphs, we can determine if there is a mathematical relationship between the two variables (e.g. is there a connection between a movie’s box office takings and the ratings which it receives in major newspapers). To get this, a program would extract the data out of the spreadsheet (CSV), rank the two variables in ascending order, and then plot this on a graph.
Sandru stated that Matplotlib could be used, as it is an attachment to Python specifically to draw graphs.
ACTION- Approach method agreed to, added to project plan.
- Future development
As there are time constraints on the project, some ideas were raised for future development including
-Login systems so that individual preferences can be recorded
-An interactive ‘how to’ help system to aid users
-A feature to enable users to store individual preferences for the recommendation engine to use.
ACTION- These suggestions amongst others are to be provided to Sandru for inclusion in the final report.
- Phase III development
Phase III of the project requires a report on a proposed recommendation engine for movies. The aim is to get as many relevant results for users. Suggestions from team members to aid this include:
-Tabris suggested the idea of allowing users to determine what factors are most important to them in watching a movie. Users would be required to answer a questionnaire on what they consider to be most important about a movie (i.e. is it the actors, director, ratings from popular literature). A current example of this would be the Flixster “application” available to Facebook users.
-The alternative to this would be to arbitrarily set the importance of different factors.
- Problems
There were a number of problems raised by the team - Browser compatibility
Since there is a tendency for web pages to be displayed differently in Internet Explorer and Firefox, the course of action decided by the team would be to code the HTML output according to the relevant standards set by the W3C. However if this causes any problems between the two browsers, compatibility with Firefox would be favoured.
- HTML interface
The problem is that the interface needs to be as user friendly as possible. A draft interface is required to gauge exactly what is required by users.
ACTION- Tabris to design a draft HTML interface
- Delegation of tasks
As all team members will be required to do Python coding, there is no need to delegate this role. However other roles have been delegated. - Tristan will focus on the mathematical side of the programming, such as identifying trends and creating mathematical equations relevant to the project. He will also managing the testing phase of the project to ensure the solution behaves in the proper fashion.
- Sandru will manage most of the administrative aspects of the project, such as organising team meetings, writing minutes and contacting Informatics staff/students when required. He will also be working on the elements of the final presentation (alongside the others) to determine how the material can be delivered in an engaging manner.
- Tabris will be focusing on the front end development of the project such as the HTML output and CSS/graphical design.
- Tasks for next meeting
Tristan- Approach section of project plan
Sandru- Polishing project plan
Tabris- Designing draft HTML interface
Important Dates
7th April- Project plan due at 9:00AM SHARP
9th April- Discussion on two dimensional data storage
