Public dashboard

[MatthewChatham]

No description provided.

[MatthewChatham]

And deploy to Heroku.

[MatthewChatham]

dash: https://appsilondatascience.com/blog/rstats/2018/05/08/dash.html?utm_campaign=News&utm_medium=Community&utm_source=DataCamp.com

[MatthewChatham]

part of speech tags: https://cs.nyu.edu/grishman/jet/guide/PennPOS.html
nltk "categorizing and tagging words": https://www.nltk.org/book/ch05.html
nltk "extracting information from text": https://www.nltk.org/book/ch07.html#tab-db-answers
ner: https://medium.com/@dudsdu/named-entity-recognition-for-unstructured-documents-c325d47c7e3a

[MatthewChatham]

http://scikit-learn.org/stable/modules/feature_extraction.html#customizing-the-vectorizer-classes

how to pass a custom analyzer

[MatthewChatham]

Ideas:

• Top words (excluding vs. including stopwords checkbox)
• Drill-down from words to articles (e.g., click "trump" bar in top words and be taken to a list of articles that feature that word in the headline). Drill down to some sort of network view that connects articles based on using the same topic/word.
• Allow people to navigate directly to any article found in drill-down

MVP: Single chart of top words, investigate D3 or something for drill-down to topic network.

[MatthewChatham]

"Deploying Dash Apps"
"Deploying Python Applications with gunicorn"
"Dash Tutorial: Part 1: App Layout"

[MatthewChatham]

Outline:

Introduction: Draw on case study and EDA, talk about headlinese
Table 1: Scrolling output of all records
Chart 1: Count of headlines by publication
Chart 2: Top words by publication
Chart 3: Headline length by publication
Chart 4: Count of top collocations by publication
Chart 5: PCA scatter + headlines in tooltip