Icon

Twitter Sentiment Predictor Data Apps Webinar

Deploying a Sentiment Analysis Predictive Model - Supervised Machine Learning

This worflow applies an RF model, trained on the Kaggle Dataset (https://www.kaggle.com/crowdflower/twitter-airline-sentiment), on new tweets around a particular airline to predict their sentiment. The last component visualizes: 1. the bar chart with the number of negative/positive/neutral tweets; 2. the word cloud of all collected tweets; 3. the table with all collected tweets. Selecting a word in the word cloud selects the corresponding tweets the word in contained into.

1. Collect and process twitter data. Users could also collect data from more brands simultaneusly and then concatenate tables. 2. Prepare and classify new data. Create the same "word vector space" as in the workflow forbuilding a sentiment predictor and represent each tweet with a document vector in the createdword vector space. Then appy the RF model.Tip: This blog post describes different document encoding options https://www.knime.com/blog/text-encoding-a-review. 3. Visualize data - bar chart of # positive, negative, neutral tweets - word cloud - table with tweets Deploying a Sentiment Analysis Predictive Model - Supervised Machine LearningThis workflow applies an RF model, trained on the Kaggle Dataset (https://www.kaggle.com/crowdflower/twitter-airline-sentiment), to predict sentiment on new tweets around the query"to:AmericanAir." This query retrives tweets directes to American Airlines. The last component visualizes: 1. the bar chart with the number of negative/positive/neutral tweets; 2. the wordcloud of all collected tweets; 3. the table with all collected tweets. Selecting a word in the word cloud selects the corresponding tweets the word in contained into.NOTE: Users can edit the query by opening the component. Common steps(e.g., lower case,stop wors, etc.)Convert strings toto documentsKaggle VectorSpace (Predictors)RF ModelEnter yourTwitter API Credentialsadd back predicted sentiment String Emoji Filter Enrichment andPreprocessing BoW andVector Space Strings To Document DocumentVector Applier Model Reader Model Reader Document DataExtractor Column Filter Random ForestPredictor Twitter Keys Joiner Twitter Query Visualizations 1. Collect and process twitter data. Users could also collect data from more brands simultaneusly and then concatenate tables. 2. Prepare and classify new data. Create the same "word vector space" as in the workflow forbuilding a sentiment predictor and represent each tweet with a document vector in the createdword vector space. Then appy the RF model.Tip: This blog post describes different document encoding options https://www.knime.com/blog/text-encoding-a-review. 3. Visualize data - bar chart of # positive, negative, neutral tweets - word cloud - table with tweets Deploying a Sentiment Analysis Predictive Model - Supervised Machine LearningThis workflow applies an RF model, trained on the Kaggle Dataset (https://www.kaggle.com/crowdflower/twitter-airline-sentiment), to predict sentiment on new tweets around the query"to:AmericanAir." This query retrives tweets directes to American Airlines. The last component visualizes: 1. the bar chart with the number of negative/positive/neutral tweets; 2. the wordcloud of all collected tweets; 3. the table with all collected tweets. Selecting a word in the word cloud selects the corresponding tweets the word in contained into.NOTE: Users can edit the query by opening the component. Common steps(e.g., lower case,stop wors, etc.)Convert strings toto documentsKaggle VectorSpace (Predictors)RF ModelEnter yourTwitter API Credentialsadd back predicted sentiment String Emoji Filter Enrichment andPreprocessing BoW andVector Space Strings To Document DocumentVector Applier Model Reader Model Reader Document DataExtractor Column Filter Random ForestPredictor Twitter Keys Joiner Twitter Query Visualizations

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: fvill
Created at: 2021-04-12
On NodePit since: 2022-10-07
Last update: 2024-04-27
Created with KNIME version: v4.6.3
Tags: sentiment analysissentimentmachine learningSVMdeploymentMarketing Analytics

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!