visualize_dataset.py

import streamlit as st
from pathlib import Path
import json
from support_functions import HealthseaSearch

def visualize_dataset():
    # Configuration
    health_aspect_path = Path("data/health_aspects.json")
    product_path = Path("data/products.json")
    condition_path = Path("data/condition_vectors.json")
    benefit_path = Path("data/benefit_vectors.json")

    # Load data
    @st.cache(allow_output_mutation=True)
    def load_data(
        _health_aspect_path: Path,
        _product_path: Path,
        _condition_path: Path,
        _benefit_path: Path,
    ):
        with open(_health_aspect_path) as reader:
            health_aspects = json.load(reader)
        with open(_product_path) as reader:
            products = json.load(reader)
        with open(_condition_path) as reader:
            conditions = json.load(reader)
        with open(_benefit_path) as reader:
            benefits = json.load(reader)
        return health_aspects, products, conditions, benefits


    # Functions
    def kpi(n, text):
        html = f"""
        <div class='kpi'>
            <h1 class='kpi_header'>{n}</h1>
            <span>{text}</span>
        </div>
        """
        return html


    def central_text(text):
        html = f"""<h2 class='central_text'>{text}</h2>"""
        return html

    # Loading data
    health_aspects, products, conditions, benefits = load_data(
        health_aspect_path, product_path, condition_path, benefit_path
    )
    search_engine = HealthseaSearch(health_aspects, products, conditions, benefits)

    # KPI

    st.info("""This app showcases a dataset of up to one million reviews that was analyzed by the Healthsea pipeline. You can search for any health aspect, whether it's a disease (e.g. joint pain) or a positive state of health (e.g. energy), the app will output a list of products and substances.
    These products have a score which is calculated by the content of their reviews.""")

    st.warning("""Please note that Healthsea is a research project and a proof-of-concept that presents a technical approach on analyzing user-generated reviews.
    The results produced by Healthsea should not be used as a foundation for treating health problems and neither do we want to advocate that supplementary products are able to solve all health issues.""")


    st.markdown("""---""")

    st.markdown(central_text("🎀 Dataset"), unsafe_allow_html=True)

    kpi_products, kpi_reviews, kpi_condition, kpi_benefit = st.columns(4)

    def round_to_k(value):
        return str(round(value/1000,1))+"k"

    kpi_products.markdown(kpi(round_to_k(len(products)), "Products"), unsafe_allow_html=True)
    kpi_reviews.markdown(kpi(round_to_k(int(933240)), "Reviews"), unsafe_allow_html=True)
    kpi_condition.markdown(kpi(round_to_k(len(conditions)), "Conditions"), unsafe_allow_html=True)
    kpi_benefit.markdown(kpi(round_to_k(len(benefits)), "Benefits"), unsafe_allow_html=True)

    st.markdown("""---""")

    # Expander
    show_conditions, show_benefits = st.columns(2)

    with show_conditions.expander("Top mentioned Conditions"):
        st.write(search_engine.get_all_conditions_df())

    with show_benefits.expander("Top mentioned Benefits"):
        st.write(search_engine.get_all_benefits_df())

    st.markdown("""---""")

    # Search
    search = st.text_input(label="Search for an health aspect", value="joint pain")
    n = st.slider("Show top n results", min_value=10, max_value=1000, value=25)

    st.markdown("""---""")
    st.markdown(central_text("🧃 Products"), unsafe_allow_html=True)

    st.info("""The products are scored based on what reviewers say. Additional variables in the scoring function are product rating, helpful count and whether the review is considered 'fake'. """)

    # DataFrame
    st.write(search_engine.get_products_df(search, n))

    # KPI & Alias
    aspect_alias = search_engine.get_aspect(search)["alias"]

    kpi_product_mentions, kpi_alias = st.columns(2)

    kpi_product_mentions.markdown(kpi(len(search_engine.get_aspect(search)["products"]), "Products"), unsafe_allow_html=True)

    if len(aspect_alias) > 0:
        kpi_alias.markdown(
            kpi(len(aspect_alias), "Similar health aspects"),
            unsafe_allow_html=True,
        )

        vectors = []
        main_aspect = search_engine.get_aspect_meta(search)
        vectors.append((main_aspect["name"], main_aspect["vector"]))
        for aspect in aspect_alias:
            current_aspect = search_engine.get_aspect_meta(aspect)
            vectors.append((current_aspect["name"], current_aspect["vector"]))
        st.markdown("\n")
        st.info("""To improve the search, the table also shows results of other health aspects with a high similarity""")
        #st.write(search_engine.tsne_plot(vectors))
        search_engine.pyvis(vectors)

    st.markdown("""---""")

    # Substances
    st.markdown(central_text("🍯 Substances"), unsafe_allow_html=True)
    st.info("""The scores of the substances are based on the products""")

    # DataFrame
    st.write(search_engine.get_substances_df(search, n))
    kpi_substances, empty = st.columns(2)
    kpi_substances.markdown(
        kpi(len(search_engine.get_aspect(search)["substance"]), "Substances"),
        unsafe_allow_html=True,
    )