MétaCan
Menu
Back to cohort
Record W3128179785 · doi:10.18280/rces.070403

Performance Evaluation of Email Spam Text Classification Using Deep Neural Networks

2020· article· en· W3128179785 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

venuePublished in a venue whose home country is Canada.
no affNo Canadian affiliation: this work is invisible to an affiliation-only frame.
No Canadian affiliation. An affiliation-only frame, the usual design, would never have seen this work. It is one of the works that make the case for inverting the frame.

Bibliographic record

VenueReview of Computer Engineering Studies · 2020
Typearticle
Languageen
FieldComputer Science
TopicSpam and Phishing Detection
Canadian institutionsnot available
Fundersnot available
KeywordsComputer scienceArtificial intelligenceSpammingForum spamMachine learningContext (archaeology)SpambotFilter (signal processing)Artificial neural networkBag-of-words modelScripting languageDeep learningInformation retrievalWorld Wide WebThe Internet

Abstract

fetched live from OpenAlex

Spam in email box is received because of advertising, collecting personal information, or to indulge malware through websites or scripts. Most often, spammers send junk mail with an intention of committing email fraud. Today spam mail accounts for 45% of all email and hence there is an ever-increasing need to build efficient spam filters to identify and block spam mail. However, notably today’s spam filters in use are built using traditional approaches such as statistical and content-based techniques. These techniques don’t improve their performance while handling huge data and they need a lot of domain expertise, human intervention and they neglect the relation between the words in context and consider the occurrence of the word. To address these limitations, we developed a spam filter using deep neural networks. In this work, various deep neural networks such as RNN, LSTM, GRU, Bidirectional RNN, Bidirectional LSTM, and Bidirectional GRU are used to a built spam filter. The experimentation was carried out on two datasets, one is a 20 newsgroup dataset, which contains multi-classes with 20,000 documents and the other is ENRON, a dataset contains 5,000 emails. The custom-designed models have performed well on both benchmark datasets and attained greater accuracy.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.919
Threshold uncertainty score0.452

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.089
GPT teacher head0.305
Teacher spread0.216 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it