Saturday, September 6, 2008

Movie Trends by plot keywords/genres

Since I wanted to do something that is related to either words or movies, I thought I could combine them and do both. I'm going to use IMDb (Internet Movie Database) as my primary resource and now I'm gathering information regarding plot keywords and genres of each movie and also the box offices. With the information I want to show the movie trend and show which plot/genre is more successful at certain time period.

Audience:
The audience would be basically anybody who is interested in movies. Showing movie trends by plot keywords could also inform movie writers and in a way suggest them what kind of plot to write next by showing them how certain types of movies are more popular.

Goal:
To inform movie makers what kinds of movie are popular among moviegoers (more likely to be successful). It's a topic that I'm also personally interested in and I think anyone who likes movies would be interested in seeing this kind of information.

Information being used:
Generally there is going to be two types of information. One is based on words (plot keywords, genres, name of the movie) and the other is based on numbers (box office, ratings). Right now, the main information that is going to be used are the genres and the box offices since I'm showing how certain genres are more successful than the other. If I can go further (still doing the research), I want to use plot keywords instead which are usually more detailed than genres.

LATCH:
Location - see whether certain genres are more successful at a certain country
Alphabetical - the plot keywords, name of the movies
Time - observe the trend throughout time or season (i.e. summer vs. winter) and see if there are any changes
Categorical - genre?
Hierarchical - Box office, ratings

Temporal Map:
Most of the visualizations that I've seen that are related to movie data were numerical and statistical graphs. I want to do something different and build a map that is based on words. Something that looks like WORDLE. The numbers, such as the $ amount of box offices will NOT be the primary information that is going to be looked at, but will "support" the verbal data and would determine what plot keywords should be highlighted, be bigger etc. I'm still exploring to see if I can integrate other variables as well.

Populations to natural diasters

I want to show the effects of natural disasters and how they affect the population growth of a country. In recent decades the growth of natural disasters average annual rate has increased to 8.4%.

3 Types of Information:
global population:
1. Fertility rates
2. Mortality rates
3. number of natural disasters
4. nation’s GDP (show their economic level)
Above are the three types of information (fertility rate, mortality rate, number of natural disasters). However I believe there is a significant factor missing, the nation’s GDP. The economic levels could reflect how that nation handles the after-effects of a natural disaster.

Organize According to Latch:
1. Fertility rates
2. Mortality rates
3. number of natural disasters
4. nation’s GDP (show their economic level)

- global population: Fertility/Mortality rates
Location – map of the globe showing the densities of rates by continent
Time – showing the change over time

- number of natural disasters:
Category - there are different kinds of natural disasters, weather, earth, fire, health, showing
Hierarchy - how intense a certain disaster is
Time – showing the change over time

- nation’s GDP:
Location - map of the globe with a color gradation showing the denser countries with more wealth


Build temporal map:
Capturing this information into a geographic map that changes from year to year. An idea of how to show he effects of natural disaster is to have two maps of the globe side to side, one that has the population growth from year to year and the other the population growth with the natural disasters. However I believe this map is not explicit enough to show the effects, I feel that there might be a better way. This only the initial idea of what I have thus far.

Friday, September 5, 2008

So far, I could find walmart's annual reports and some of its competitors' like target, costco, kmart (a little dated though) and some websites where I could read about some controversial issues regarding walmart, such as wages, its impacts on local labor markets, workers' health care etc. I haven't really digged into those resources yet, but was trying to structure a big picture of my design to get started.

Audience
It's not specific to a certain group but rather to general public, because I think walmart can interest anyone who's ever shopped there.

Goal
To help people become aware of walmart's presence in and impacts on domestic + global economy and have critical eyes on them.

Groups of Information
There will be two types of information: number-based factual data and text-based issues and arguments (I'm thinking of adding some stories about, say, a family shopping at walmart or a labor worker who's not properly paid, so that the audience can relate themselves to the info more easily, but not sure yet.)

Temporal Map
First, facts on the magnitude of walmart (net sales, growth rate, # of stores and employees, etc.)
: This part should give the audience more concrete ideas how gigantic walmart's presence is.
Second, facts on low wages behind low prices, job opportunities sent overseas, etc. and the arguments brought up against them.
: This part should encourage the audience to think critically about the issue.
(I'll be sub-dividing these two big parts as I proceed.)

LATCH
Location for where the stores are globally
Time for walmart's growth over time
And I'll be most likely using Categories to organize the info. I haven't gone to the details yet.

This is my idea as of now. I feel like maybe I'm trying to take on too broad an issue, but as I get more familiar with my data and sources, some of them mentioned above could be taken out making my focus/goal more specific.

Thursday, September 4, 2008

The Plan

For the record, it’s really hard for me to do this without visualizing it first. Below is how I see it in my head at the moment.

As tempting as it is to do a location-based approach, I feel like this will be a glorified version of http://www.suicidology.org/associations/1045/files/2005HealthRegionsSlide.pdf and I’m not all that interested in redesigning this map. I think I’d like to try something different and organize my information hierarchically. A block will represent each state or something similar and then re-organized according to user criteria. Also, based on how states in particular categories, their representations could grow or shrink. So I guess LATCH wise, I’m thinking Hierarchy, Category and then comparisons of scale.

I also am going to show all 52 states all the time (just not in map form), so people can have multiple points of reference. This may be subdivided by 20% increments, so it’s easy to visually distinguish where each state falls.

Monday, September 1, 2008

Another idea

I was trying to think of other ideas and came across one.

I thought it might be interesting to look at search trends in different countries. Search engines have become essential part of our internet experience, and billions of searches made every day reflect needs and interests of users all around the world. From looking at search trends at specific time(s) of the year, it might be able possible to map and compare different trends (e.g., cultural, political, economical, etc) between different countries. Current events can be combined to see how they affected people in different countries.

There are sources like Google Insight for Search and Google Trends that provide attributes such as
  • Different countries' interests for a specific topic (can be narrowed down by specific time range)
  • Rising searches and their popularity growth rate
  • Graph for the interest on a specific topic and related news over a timeline
I am feeling like I would need to choose a set of global topics/issues and get relevant data for a set of countries.

I am wondering if this is an interesting&difficult enough topic to pursue. Any feedback will be really appreciated. Thank you.

my ideas

I think in this time and age the effects of population growth will play such a huge factor of what we do in our day to day lives. And what we do now will be paramount to what will be done in the future. However in order to do anything about it in the present, people need to be aware and knowledgeable the large increase of the population growth. I want to show those who may not of the time and the accessibility to this knowledge the future effects of today's population growth.
As a design exercise, I also want to combine the relationships amongst different visualizations. Such relationship as:
  • population growth to pollution changes
  • population growth to economic trends (gas prices, inflation)
  • historical change in voter demographic
Below are some attributes:
  • Showing geographic regions
  • demographic change in population
  • life expectancy, # of births
Perhaps there will be hidden trends amongst the different relationship. If you have any other ideas on relationship or attributes please let me know :). THANK YOU!

Initial ideas

When I first heard about the assignment, I started to find examples from my everyday life and thought about what information I would like to see visualized.

I'm sure this kind of info visualization is popular, but I'm interested in showing data about words and languages. I like to read in my leisure time and even reading just one book, I realize certain words appear more frequently. Sometimes it has to do with the genre and context of a novel (i.e. "robot" if it's a sci-fi novel, the name of the character, etc.) and sometimes it simply has to do with the language it is written in. So my initial thought was visualizing the following:

- words that most frequently appear in books in the US
- words that most frequently appear in books by genre (sci-fi, romance, etc.)
- words that are most frequently used/spoken in the world (by country)
- words that most frequently appear in songs
- I think somebody has already done this, but maybe showing which words appear most frequently in a Bible (if my memory is correct, it wasn't "God")

I think the challenge I'll be facing is collecting reliable data for specific topics (i.e. books by genre, by country, poems, songs...). There is a website called WORDCOUNT where I can see what words are more frequently used (written & spoken), but only in English. I think observing frequently used words tells a great deal about culture, so I'm interested in knowing the word count for other countries as well. (Anybody know the source?) I'm generally interested in international data, but the problem here is that I don't know how to define "international" (in other words, how many countries I'll be able to include when collecting data).

Another topic I'm very interested in is movies. I'm a huge fan of movies and do a lot of research on them before going to the theater. I'm therefore interested in finding ways to make the process more fun & easy for moviegoers. I realized couple of people are already doing their research on movies, so I'm not sure if I'm allowed to do it?

Hope you can give me some advice.
Thanks in advance :)