Tuesday, November 15, 2011

Reaction: Balancing Systematic and Flexible Exploration of Social Networks

The author in this paper builds a tool called Social Action and uses it to analyze social networks more systematically. The work is based on lot of previous work in network visualization. I feel that the analysis of network is a research topic that requires continuous innovation and development. It is of high importance to a social being now a days and all the industries are revolving around it. It also states that the tools that were developed earlier for the same purpose were compromising in some of the features in a total list of them while include flexibility in navigation, visualization of connectivity, aggregating the number nodes to display when the size of network becomes large etc. This tool is aimed at accommodating all such features.

The aggregation of the nodes makes visualization simpler and more legible. The ranking of the resulting groups aid the user in visualizing the relationships of the user more closely. But this operation may be computation intensive. I feel more information on it should have been given. Thus, this tool helps in visualizing network more effectively and flexibly.

Social Action uses attributed ranking and coordinated views to achieve that which is done using ordered lists, color coding, dynamic queries etc. The highlighting of some areas of the visualization based on the input given by the user is also of great practical importance. I think most of the present day social networking sites do this to make the webpage look more customized to the user.

The main advantage of this tool is, based on the relationship the user is interested in, it has the capability to visualize data in different ways. The tool increases the legibility of the visualization by masking some of the data that is of least significance or of least relevance. The ability of the tool to process data based on different criteria really interests me, but clarity on the computational intensity would have made things more clear.

The implementation zooming in and out feature interested me as it is very difficult to implement it as the state in which user is at might be difficult maintain, ending up is creation of confusion in user's mind as discussed in class. I completely agree by his statement that it makes visualization manageable, but am a bit skeptic about his statement that it makes visualization easy to interpret. There is no hint of the comparison of this tool with its peers in terms of complexity, which kind of makes me think that this may be a badly performing tool compared to the rest. More insight should have been given in this dimension.