Classifying SMS as Spam or Ham: Leveraging NLP and Machine Learning Techniques
Bibliographic record
Abstract
In an era dominated by mobile communication, Short Message Service (SMS) plays a pivotal role in interpersonal interactions.However, the surge in unsolicited spam messages necessitates effective differentiation mechanisms.This exploratory data analysis (EDA) utilizes a dataset from the renowned UCI Machine Learning Repository to discern key characteristics distinguishing spam from legitimate messages.Employing Natural Language Processing (NLP) technique vectorization (BOW and TF-IDF), including the use of a Naï ve Bayes algorithm and sentiment analysis, this investigation uncovers patterns and peculiarities specific to spam content.The findings highlight distinct differences in lexicon usage, message structure, and linguistic markers between spam and legitimate messages.For instance, spam messages often exhibit aggressive language and utilize unconventional structures.To elucidate, specific examples of such language patterns and structural anomalies are provided, offering a more nuanced understanding of the study's outcomes.Rooted in data-driven insights, this study lays the foundation for future endeavours in developing robust, NLP-powered spam detection mechanisms to preserve the essence of personal communication in the SMS sphere.Evaluating the model on a test dataset of 5,572 SMS messages yielded noteworthy results.The model demonstrated a precision rate of 98% for legitimate messages and an impeccable 100% precision for identifying spam without any false categorization.However, a notable dip in the recall rate for spam messages, at 85%, raises important considerations.This suggests potential challenges in detecting certain types of spam, emphasizing the need for further refinement in the model.The respective f1-scores for ham and spam messages were 99% and 92%, shedding light on the model's overall efficacy.These performance metrics not only quantify the model's accuracy at an admirable 98% but also prompt deeper reflections on the practical implications of the results, emphasizing areas for future research and enhancement in spam detection mechanisms within the dynamic landscape of mobile communication.
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.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot 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.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".