highlighter.py

from lime.lime_text import LimeTextExplainer
from nltk.tokenize import sent_tokenize
from predictors import predict_for_explainanility


def explainer(text, model_type):
    def predictor_wrapper(text):
        return predict_for_explainanility(text=text, model_type=model_type)

    class_names = ["negative", "positive"]
    explainer_ = LimeTextExplainer(
        class_names=class_names, split_expression=sent_tokenize
    )
    sentences = [sent for sent in sent_tokenize(text)]
    num_sentences = len(sentences)
    exp = explainer_.explain_instance(
        text, predictor_wrapper, num_features=num_sentences, num_samples=500
    )
    weights_mapping = exp.as_map()[1]
    sentences_weights = {sentence: 0 for sentence in sentences}
    for idx, weight in weights_mapping:
        if 0 <= idx < len(sentences):
            sentences_weights[sentences[idx]] = weight
    print(sentences_weights, model_type)
    return sentences_weights, exp


def analyze_and_highlight(text, model_type):
    highlighted_text = ""
    sentences_weights, _ = explainer(text, model_type)
    min_weight = min(sentences_weights.values())
    max_weight = max(sentences_weights.values())

    for sentence, weight in sentences_weights.items():
        normalized_weight = (weight - min_weight) / (max_weight - min_weight)
        if weight >= 0:
            color = f"rgba(255, {255 * (1 - normalized_weight)}, {255 * (1 - normalized_weight)}, 1)"
        else:
            color = (
                f"rgba({255 * normalized_weight}, 255, {255 * normalized_weight}, 1)"
            )

        sentence = sentence.strip()
        if not sentence:
            continue

        highlighted_sentence = (
            f'<span style="background-color: {color}; color: black;">{sentence}</span> '
        )
        highlighted_text += highlighted_sentence

    return highlighted_text