Commit d8744a9d by Sacha Committed by SXibolet@2PITAU

### closed #9 and some other tweaks

 ... ... @@ -131,9 +131,10 @@ def logout(): @application.route("/register_anon") def register_anon(): u = session['username'] session.pop('username', None) session.pop('user', None) u = session.get('username') if u is not None: session.pop('username', None) session.pop('user', None) return redirect('/register') @application.route("/account", methods=['GET', 'POST']) ... ...
 see spec here __
 \documentclass[a4paper,12pt]{article} \begin{document} \pagestyle{empty} \begin{enumerate} \item Switch to conditional probability for linguistic features: Recall $p(X_{sent} = x_i) = \frac{P(x_i) \in C'_i}{\sum_{i=1}^{z} f(x_i) \in C'_i}$ is equivalent to the number of comments (tweets) with sentiment $x_i$ divided by the total number of comments. Therefore, $p(X_{caps} = x_i) = p(X_{sent} = x_i | C'_i \in Caps) = \frac{p(X_{sent} = x_i \cap C'_i \in Caps)}{p(C'_i \in Caps)}$ where $Caps$ is the set of comments that have capitalized terms. This works similarly for the extreme lexicon. With SentiStrength, sentiment scores range from $-4$ to $4$ thus $z=9$. \end{enumerate} \end{document}
 pdflatex fa15-report.tex open fa15-report.pdf
This diff is collapsed.
 ... ... @@ -20,7 +20,7 @@
 ... ... @@ -41,7 +41,7 @@

{{ form.hidden_tag() }} {{ form.email(value=username, type="email", class_="field", placeholder="email (username)", required="required") }} ... ...
 ... ... @@ -39,7 +39,7 @@