This week we have settled down our project direction. We met our client on Monday and presented our two ideas to him. Our decisive client immediately chose the one that provides a bigger challenge for us, which aims to analyze data about game mods (we will explain later) for Sims 4.
Since Sims is a game about virtual life, it has a big modding community. After our research, we divided the mods into three types: customization mods, behavior mods and gameplay mods. Customization mods are the things you would like to ‘physically’ customize in the game; hair, clothing, furniture and houses. It covers the biggest percentage of all the mods, since people want more customization in the game and it those mods are relatively easy to create. The behavior mods are other things you add to the game. For example, a new trait mod. The last type is the gameplay mods, which “fix” the game by changing its main mechanism. For example, rewriting the social system of Sims 4. The later two kind we briefly call them game mods. Although these two types only take up a small percentage, our client is interested in seeing more detail about them in order to see and understand what the community is changing.
Our goal is to gather information of game mods from the most popular Sims modding sites. After we settled our goal, we did more research about the websites, the game mods, and drafted questions that could be answered by further analysis.
There are several challenges for us. First, there are currently no good categorizations for game mods. The existing categorization is based on how these mods are installed in the game. We will analyzing the text and using keywords to figure out a better way to categorize the content for our client. Second, the data set is fairly small which means we only have a limited number we can gather information from. In addition, the websites themselves are not well organized. There are some inconsistencies and inaccuracies in the data. For example, on one website there are two places showing the downloads, but the two do not match. Also, a lot of the mods information are input from the uploader, and the upload process is a little confusing which results in some inaccuracy. The keywords are not reliable enough, and the mod’s dependency on packs are occasionally wrong. We have done a lot of research and brainstormed about what questions we can answer from these sets of data.
Along with our research, we have started prototyping so that we can see the rough data and set a baseline for further analysis. Also we want to save the data in case we need to analyze the mod’s life cycle in the future. The data will be the existing stats of each mods, including upload time, downloads, views, likes, etc.
Next week, we will present our prototype to the client and hopefully we can interview some data analysts during the quarter walk around. Looking forward to it!