Publications by Isha Akshita Mahajan
Document
In the previous I ran a a structural topic model with K = 6. This was deciphered by using the searchK() function and looking at the coherence values for statistical fit. For meta data, I used organization so I can decipher which words occur in each publication. As seen in in the previous visualization, topic 2 seemed to be the most prevalent in a...
1357 sym R (9667 sym/75 pcs) 7 img
Post 6
From the clusters, I was able to see which topics have similarities at the same height. There were three distinct clusters that came out, and the word clouds help add context to the clusters that were found in the previous post. The advantage of a structural topic model that I was able to observe the topics using the covariates as the organizatio...
731 sym
blog post 4
Structural Topic Model I decided to work with structured topic modeling for this analysis. This is because there is significant meta data available in these articles, for example the date they were published, the organized they were published in etc. While I don’t work on incorporating the metadata in this analysis, there is space in this analy...
907 sym R (54841 sym/21 pcs) 4 img
Post 3
Tokenization + Preprocessing Analysis on this data was done using the Quanteda package in R. After manually selecting articles published during and after the year 2000, I began pre-processing the text. First, I created tokens by word which reduce the text into smaller and more interpretable objects. Next, I perform a series of pre-processing step...
1474 sym R (1632 sym/9 pcs) 1 img
Post 1
Motivation Throughout my time at university, I have been passionate about exploring topics in international affairs. During my sophomore year, I enrolled in a International Journalism class, and my professor Shaheen - a bold woman of journalistic integrity mentored me to explore media landscapes through an expansive lens. As I now near the comple...
2073 sym
Blog post 2
Data and Methods I will be using data from a dataset created by Farris et al. for this project. It represents a compilation of over 14,000 sources of human rights texts from four sources - Amnesty International, Human Rights Watch, Lawyers Committee for Human Rights and the US Department of State. In our world where media narratives are fragile ...
1155 sym