Social Networking Course Reflection

Coursework

Besides gaining theoretical knowledge about information, human cognition, social networks and social media, human centric design and analytics, etc. through attending lectures, there are also several assignments for us to have a more comprehensive understanding of this Social Networking course. In blog-post assignments, I have built my own blog, wrote 4 blog posts in total involved in social media analytics and Sustainable Development Goals (SDG) areas and exchanged ideas by giving comments to my classmates’ blogs. Then, I applied my 2 programming assignments on my own blog with 2 different analytic methods including sentiment analysis and social network analysis in order to see a big picture of my online social network objectively.

Analysis Results

1. In programming assignment one: Sentiment Analysis…

Figure 1 Sentiment Analysis

I search for web crawler method and wrote a python program to collect all the comments I received from my blog in a .txt file, which is my first-time crawling. Then I wrote another python program with NLTK python package to perform sentiment analysis on the comments using the dictionary-based approach and get the output. I have received 18 comments with 1667 words in total, among which are 76 positive words and 26 negative words respective. The overall sentiment score (calculated by formula i) of my received comments is around 0.03. As the score is a positive number, it is reasonable to draw a conclusion that the polarity of my received comments is positive.

2. As for programming assignment two: Social Network Analysis…

I wrote a python program with NetworkX python package base on a provided class blogging network sociomatrix to figure out my interaction and position in the class blogging network. There are totally 75 students in this class. The value of the in-degree and out-degree of the node representing me is 9 and 13 respective, which means I have received comments from 9 different classmates and provided comments to 13 classmates. Closeness centrality is based on the inverse of the geodesic distance of the actor to every other actor in the network, which increases with decreasing distance between actor to other nodes. The value of the closeness centrality of the node representing me is about 0.44, showing that my closeness centrality and distance to other classmates is in the middle level in the whole class. Betweenness centrality is the average proportion of the actor occupies the shortest path of any other two nodes. The value of the shortest-path betweenness of the node representing me is about 0.033, this not that high value reflecting that I did not play a center role in our class blogging network.

Figure 2 Class Blogging Network Sociomatrix

Prospects

This course led me to the social networking field and inspired my enthusiasm on social media analytics. After this course, I will still pursue more knowledge about social media analytics and do more practice to consolidate my cognition in this technical area.