About Us

What is Phishing Data Insights?

This site presents Spring 2021 DSCI 550 student work on phishing and fraudulent email data. The assignment asked teams to turn phishing datasets into a web data visualization site: extract and summarize the raw data, convert the results into JSON/CSV/TSV for D3, build browser-based visualizations, and connect the work to larger-scale search and exploration tools such as Solr, ImageSpace/ImageCat, and GeoParser.

Students explored how fraudulent messages vary by time, sender geography, subject and body vocabulary, urgency, attacker titles, social-engineering cues, language style, and pairwise similarity. The local archive includes the Kaggle Fraudulent E-mail Corpus used for fraudulent message analysis, plus a copy of Phishing_Legitimate_full.arff, which corresponds to Choon Lin Tan's Mendeley phishing feature dataset.

Data Integration & Search

Connecting phishing email content, sender metadata, inferred geography, and extracted keywords into queryable datasets for visualization and search.

Interactive Visualizer

D3.js charts, heatmaps, bubble charts, histograms, and circular bar plots created from the enriched fraudulent-email dataset.

Publicizer

Student-built phishing data mini-sites and search artifacts that demonstrate the educational value of cybersecurity datasets and modern data science tooling.

Student Visualization Teams

Team 4

Team 4 investigated phishing attack behavior across time, geography, urgency, wording, and language style. Their work includes calendar frequency views, attacker-location maps, time-of-day urgency bars, phishing and reconnaissance word clouds, and a sunburst of misspellings and capitalization patterns.

View Team 4 Visualizations

Team 8 / Team Banana

Team 8 enriched fraudulent email records with timestamps, sender context, social-engineering tags, attacker titles, and similarity scores. Their D3 views show sent-time signatures, common phishing terms, attacker title frequencies, unemployment context, and Tika-Similarity distributions.

View Team 8 Visualizations

Open the full team gallery.

Search & Explore

The visualizations in this gallery come from the Spring 2021 DSCI 550 phishing web data visualization assignment. Student teams analyzed fraudulent email and phishing feature datasets, extracted social-engineering and sender features, and produced D3.js visualizations plus search and ImageSpace artifacts.

Browse the Student Visualizations

Open Fraudulent E-mail Corpus on Kaggle Open Phishing Feature Dataset

Originally parts of the submitted workflow used local preprocessing, Solr ingest, and ImageSpace/ImageCat artifacts. The D3 visualizations here are static browser assets that can be previewed with a local HTTP server.

ImageCat (from the chrismattmann/imagecat repo) is an OODT/RADIX app that extracts metadata and OCR text via Tika + Tesseract, then loads records into Solr for search.
ImageSpace (from the nasa-jpl-memex/image_space repo, part of the DARPA MEMEX program) builds on ImageCat to support visual browsing, search, and similarity exploration using metadata, OCR, and computer-vision plugins.

USC Data Science Partner Sites

IRDS

The Information Retrieval and Data Science Group’s (I.R.D.S.) mission is to research and develop new methodology and open source software to analyze, ingest, process, and manage Big Data and to turn it into information.

We have expertise in data collection and contribute to the world's largest and most often downloaded open-source projects, working with NASA, DARPA, DHS, NIH across a number of domains, Earth Science,Planetary Science, Astronomy, defense, and private industry.

Visit IRDS

Credits

Dr. Chris Mattmann - Visit his website

DSCI 550 Spring 2021 Class - phishing web data visualization assignment