From ba7d2726d1cffd7e1e1b7d8ff6e030f8ad64c7c5 Mon Sep 17 00:00:00 2001 From: Jeffrey Nappi Date: Tue, 29 Apr 2014 16:23:46 -0700 Subject: [PATCH] Fixed long-form article issue w/ calculate_best_node See https://github.com/jiminoc/goose/blob/master/src/main/scala/com/gravity/goose/extractors/ContentExtractor.scala#L227 --- newspaper/extractors.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/newspaper/extractors.py b/newspaper/extractors.py index 7564c399..c8f24382 100644 --- a/newspaper/extractors.py +++ b/newspaper/extractors.py @@ -568,7 +568,7 @@ def calculate_best_node(self, article): if (nodes_number - i) <= bottom_negativescore_nodes: booster = float(bottom_negativescore_nodes - (nodes_number - i)) boost_score = float(-pow(booster, float(2))) - negscore = -abs(boost_score) + negative_scoring + negscore = abs(boost_score) + negative_scoring if negscore > 40: boost_score = float(5)