underfull \hbox (badness 1097) in paragraph

underfull \hbox (badness 1097) in paragraph in lines 190--194

*copied lines 188 to 194 (including 194)

\paragraph{Data Cleaning and Pre-processing for LDA}
% TODO_7: Alyssa: flesh this out
While there is some flexibility regarding the format of the dataset (the system currently supports .xls, .xlsx, and .csv), it is nonetheless necessary that it at least contain a column labeled as ``Example''. This column will hold the text examples, with one document or, broadly speaking, textual entity, per row.
% DT: These preprocessing steps are actually baked in. The user has no control.
%When the user uploads their spreadsheet, they are able to select the preprocessing that is appropriate for their dataset. The user is additionally able to specify parameters for some of the preprocessing steps, such as common phrase removal and multi-word entity joining. The system does all of the preprocessing outlined in below by default, but the user can toggle specific processes on and off according to their needs.
LDA employs a bag-of-words model, where each document is understood as an unordered collection of words; to make analysis more conducive to the discovery of useful topics, the system filters out extremely common and extremely rare words. The pre-processing steps we employ include the following:
\begin{itemize}