Icon

04.Callee Deploying Sentiment Predictor - Lexicon Based

Deploying Sentiment Analysis Predictive Model - Lexicon Based Approach

This worflow applies the lexicon based approach on new tweets 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, and (3) the table with all collected tweets.

If you use this workflow, please cite:
F. Villaroel Ordenes & R. Silipo, “Machine learning for marketing on the KNIME Hub: The development of a live repository for marketing applications”, Journal of Business Research 137(1):393-410, DOI: 10.1016/j.jbusres.2021.08.036.

6. Get Original Data and Joined Sentiment Predictions.

Here we execute a segment of the building workflow that joins the original data with joined predictions, when they exist. Recall that if a tweet/document does not have any sentiment word, it will have a neutral sentiment prediction associated with it. The quality of these predictions can then be evaluated by using this dataset in a web application or in a web service. Here, we send the output back for testing purposes.

5. Calculate a Sentiment Score based on the Number of Positive and Negative Words and Classify Documents based on the Score.

The sentiment score is calculated by (number of postive words - number of negative words) divided by (number of postive words + number of negative words). If the score is negative it is classified as negative, if the score is positive it is classified as positive, and if it is equal to 0 it is classified as neutral.

4. Count the Number of Positive and Negative Words per Document.

Here we re-use a shared component also present in the building workflow. It encapsulates the counting of sentiment words per document, separated by class.

3. Use Text Mining to Tag Words with Positive and Negative Meaning based on a Dictionary.

Here we execute another captured segment of the building workflow -- this time, to tag words based on their sentiment. Non-tagged words get filtered out in the end.

2. Data Manipulation/Preparation.

Here we execute a captured workflow segment of the building workflow that performs some data preprocessing. The most important node in this segment is "Strings to Document", which formats several string columns into a single document that can be text-mined in KNIME.

1. Receive text as input for sentiment prediction.

Here we receive documents (e.g., tweets) to predict their sentiment. The only requirement is that the documents table should have a column named 'text'.

Callable Version of "Deploying Sentiment Analysis Predictive Model - Lexicon Based Approach"

This callable version of our deployment workflow is useful for application testing, among other things.

Workflow Input
Numbers of Positive and Negative Words per Tweet
Data Preparation
Workflow Output
Calculate Scores
Join Sentiment Predictions and Original Data
Tag Words as Positive or Negative

Nodes

Extensions

Links