[QnAMaker] Active learning low score variation multiplier value (#361)

Active learning low score variation multiplier value
- PreviousLowScoreVariationMultiplier to 0.7
- MaxLowScoreVariationMultiplier to 1.0

Author: gurvsing

libraries/botbuilder-ai/botbuilder/ai/qna/utils/active_learning_utils.py

@@ -7,8 +7,8 @@
 from ..models import QueryResult
 
 MINIMUM_SCORE_FOR_LOW_SCORE_VARIATION = 20.0
-PREVIOUS_LOW_SCORE_VARIATION_MULTIPLIER = 1.4
-MAX_LOW_SCORE_VARIATION_MULTIPLIER = 2.0
+PREVIOUS_LOW_SCORE_VARIATION_MULTIPLIER = 0.7
+MAX_LOW_SCORE_VARIATION_MULTIPLIER = 1.0
 MAX_SCORE_FOR_LOW_SCORE_VARIATION = 95.0