Главная
Study mode:
on
1
Intro
2
Agenda
3
Feedback collection through surveys
4
How can NLP help?
5
What do you expect from the answers?
6
What options do we have?
7
LDA result on hotel reviews
8
Reasons for low coherence of LDA results
9
Alternative approaches
10
Semantic relationships in vectors
11
Semantic Clustering
12
Topic Coherence Improvements
13
Topic Distribution Improvements
14
Improved Results
15
SENTIMENT ANALYSIS
16
Examples
17
Challenges
18
Neural Network Based Approaches
19
Convolutional Neural Network
20
Recurrent Neural Network/LSTM
21
Sentiment Results
22
Other related problems Opinion Extraction
23
Learnings
Description:
Explore the latest developments in natural language processing for feedback analysis in this 42-minute conference talk from Devoxx. Dive into topic detection and sentiment analysis techniques used at Qualtrics to revolutionize text analytics. Learn about word embedding as a foundational concept for neural network models. Discover popular neural network-based approaches for topic detection and sentiment analysis, and gain insights into productizing research models for real-world applications. Examine the challenges of traditional methods like LDA and explore alternative approaches such as semantic clustering. Understand the improvements in topic coherence and distribution. Investigate sentiment analysis examples, challenges, and neural network-based solutions including Convolutional Neural Networks and Recurrent Neural Networks/LSTM. Gain valuable learnings on opinion extraction and related problems in the field of NLP-driven feedback analysis.

NLP in Feedback Analysis

Devoxx
Add to list
0:00 / 0:00