t-SNE Visualisation of public trust Patterns from the Indicators response speed and press release frequency

Project info

• Field of studyCommunication and Social Media
• LevelDoctoral dissertation (S3)
• Method typeDimensionality Reduction
• Estimated build time± 7 working days
• Toolsscikit-learn TSNE, PCA, StandardScaler
• PriceUS$31 - US$203

Research goal

Visualise the structure of public trust data in two dimensions from response speed and press release frequency with t-SNE and examine its cluster patterns.

Variables

Example result

• 0.388KL divergence
• 0.978Trustworthiness t-SNE
• 0.916Trustworthiness PCA
• 0.875Silhouette t-SNE
• 0.793Silhouette PCA
• 30Perplexity

t-SNE maps 8 indicators to 2 dimensions with trustworthiness 0.978 (PCA 0.916) and silhouette 0.875, making the groups visually distinct.

Computed from illustrative synthetic data, not from a real study.

About t-SNE for Visualisation

t-SNE maps high-dimensional data to two dimensions while preserving local neighbourhoods. It is used only for exploratory visualisation of cluster structure, not for statistical testing.

Steps

1. 1Standardise features and, if dimensionality is very high, pre-reduce with PCA
2. 2Choose perplexity (for example 5 to 50)
3. 3Run TSNE with a fixed random_state
4. 4Try several perplexities to test pattern stability
5. 5Colour points by class or cluster
6. 6Interpret with care

Assumptions to check

• Distances between clusters and cluster sizes in the plot have no quantitative meaning
• Results depend on perplexity and initialisation
• New data cannot be projected without refitting

Reading and reporting the results

Present t-SNE as a supporting illustration and state the perplexity and seed used. Do not draw conclusions about inter-cluster distances from this figure.

Example Python code

A short starting script with this project's variable names. The delivered version is adapted to your data and fully commented.

import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import make_blobs
from sklearn.manifold import TSNE
from sklearn.decomposition import PCA
from sklearn.preprocessing import StandardScaler

X, public_trust = make_blobs(n_samples=400, n_features=20, centers=4, cluster_std=3.0, random_state=22)
Xs = StandardScaler().fit_transform(X)
response_speed = Xs[:, 0]
press_release_frequency = Xs[:, 1]

Xp = PCA(n_components=10, random_state=22).fit_transform(Xs)
fig, axes = plt.subplots(1, 3, figsize=(14, 4))
for ax, perp in zip(axes, (5, 30, 60)):
    emb = TSNE(n_components=2, perplexity=perp, init="pca", random_state=22).fit_transform(Xp)
    ax.scatter(emb[:, 0], emb[:, 1], c=public_trust, s=10)
    ax.set_title(f"perplexity = {perp}")
    print("perplexity", perp, "KL divergence not exposed; embedding shape:", emb.shape)
plt.tight_layout()
plt.show()

Packages and prices

Choose how much help you need. Each higher package includes everything in the lower ones. Prices exclude 11% VAT.

• Payment: 50% down payment before work starts, balance after the deliverables are handed over
• Tax: Prices exclude 11% VAT
• Data and academic integrity: Results are computed from your own research data; you write and stand behind your manuscript, we help with the analysis and your understanding of it
• Revisions: Up to 2 revisions are included in the price (Package 3: 3 rounds)

Frequently asked questions