--- title: "Aged to Perfection — paper by Tang (2025) — summary and ideas" description: "This study asks how spoken British English varies with speaker age, and whether a speaker's age group can be… 0.73 accuracy. 7 references." authors: ["MingZe Tang"] year: 2025 venue: "arXiv:2506.17708" source: "https://arxiv.org/abs/2506.17708" doi: "10.48550/arxiv.2506.17708" arxiv: "2506.17708" licence: "CC BY-SA 4.0 (arXiv)" licence_url: "http://creativecommons.org/licenses/by-sa/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/tang-2025-aged-perfection-machine-learning-maps/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/tang-2025-aged-perfection-machine-learning-maps/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2025](https://md.research.elamur.ai/free/en/papers/2025/page/16/) / Aged to Perfection: Machine-Learning Maps of Age in Conversational English # Aged to Perfection: Machine-Learning Maps of Age in Conversational English **Authors:** MingZe Tang · **Year:** 2025 · **Venue:** arXiv:2506.17708 · **Licence:** [CC BY-SA 4.0 (arXiv)](http://creativecommons.org/licenses/by-sa/4.0/) **Links:** [Publisher's page](https://arxiv.org/abs/2506.17708) · [PDF](https://arxiv.org/pdf/2506.17708) · [DOI 10.48550/arxiv.2506.17708](https://doi.org/10.48550/arxiv.2506.17708) · [arXiv 2506.17708](https://arxiv.org/abs/2506.17708) · [OpenAlex](https://openalex.org/W6910498543) **Citations:** 0 citations (OpenAlex, 2026-10-14) ## Summary This study asks how spoken British English varies with speaker age, and whether a speaker's age group can be predicted from what they say. It uses the spoken BNC2014, which contains about 1.25k informal conversations recorded in 2012–2014 with speakers aged 2 to 91. Speakers are sorted into five cohorts: Kids, Teens, Young Adults, Adults and Seniors. The descriptive side combines ANOVA on utterance length and lexical diversity, a chi-squared search for words typical of each cohort, and BERT embeddings with UMAP and K-means. The predictive side trains five scikit-learn classifiers on speaker-level text. SMOTE handles class imbalance, and four feature families are compared. A Linear SVM on TF-IDF features reaches about 0.73 accuracy. Adding handcrafted lexical, syntactic and discourse features does not improve on that. Errors cluster between neighbouring cohorts, and t-SNE shows a smooth age gradient. The takeaway is that age shapes conversation gradually rather than in discrete generational steps, and that simple text representations already carry most of the demographic signal. ## Key points - Data: BNC2014 spoken component with 1,251 dialogues, a mean of ≈941 tokens and ≈3 speakers per dialogue. Turns are short (mean 8.76 tokens, median 5). Higher social grades and graduates are over-represented, and the median speaker age is 34. - Mean utterance length grows with age, from roughly 6 words in children to about 10 for speakers over 20 (F(4,518)=2.97, p=.019). Gender has no significant effect. - Vocabulary overlap between interlocutors drops as their age gap widens (r = −0.219). The mean age of a dialogue's speakers shows no such link. - Classification setup: - TF-IDF capped at 15,000 features, minimum document frequency of 3, stop words removed. - Seed fixed at 42. - Classification results: - The best model is Linear SVM on text features, with accuracy 0.7264 and F1 0.708. - Combining all feature families gives 0.7075. - Lexical-only features score 0.30–0.40. - Discourse features reach up to 0.62 with the MLP. - Embedding clusters (two clusters found via silhouette score) do not split by age and are dominated by markers like *like, yeah, erm*. Cosine similarity is nevertheless higher between adjacent cohorts than between distant ones. ## Interesting ideas - Run ablations by feature family against a plain bag-of-terms baseline, so you can check whether engineered features add any signal the simple representation lacks. Here they added none. - Identify the characteristic words of each group in two independent ways (chi-squared tests and logistic-regression weights) and trust only what both methods agree on. - Treat age as an ordered, continuous variable rather than unordered bins, and read the confusion matrix for errors at neighbouring boundaries versus distant ones. - Measure lexical alignment between conversation partners and relate it to how socially distant they are, a cheap dialogue-level signal. - Spoken data is saturated with hesitations and discourse markers that blur semantic clustering. ## For Mova How these ideas could serve Mova, our small open language model — its files, stages and tests: - **I1** (learn): Any new hand-built feature for the perceptron parser or tagger could be required to beat a trivial baseline in a controlled ablation, which works as a negative control against feature bloat. - **I2** (learn): This mirrors witness agreement for silver data, and it could be used to build register or speaker-style lexicons that need consensus from separate statistical witnesses. - **I3** (learn): Evaluation of graded labels (formality, certainty, intensity) could score near-misses differently from gross errors, giving more honest error analysis. - **I4** (frames): Overlap of vocabulary between speakers could serve as an explainable feature when modelling dialogue participants, for example in tracking who is accommodating to whom. - **I5** (tokens): Tagging fillers and discourse markers explicitly at the token level would keep them from distorting the UD tree and downstream frames, while still preserving them as pragmatic evidence. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* 📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works 🧭 Topics (2): [English](https://md.research.elamur.ai/free/en/topics/english/) · 2,263 works · [🗨️ Dialogue systems](https://md.research.elamur.ai/free/en/topics/dialogue-systems/) · 624 works ## Cite Cite as: MingZe Tang. 2025. Aged to Perfection: Machine-Learning Maps of Age in Conversational English. arXiv:2506.17708. The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @misc{tang2025aged, title = {{Aged to Perfection: Machine-Learning Maps of Age in Conversational English}}, author = {MingZe Tang}, year = {2025}, doi = {10.48550/arxiv.2506.17708}, eprint = {2506.17708}, archivePrefix = {arXiv}, url = {https://doi.org/10.48550/arxiv.2506.17708} } ``` - [← previous Accusation, Anger, and Defense: Rhetorical Questions in Genesis, Exodus, Numbers, and Judges](https://md.research.elamur.ai/free/en/papers/nabulsi-2025-accusation-anger-defense-rhetorical-questions/) - [next → AI-Driven Generation of Old English: A Framework for Low-Resource Languages](https://md.research.elamur.ai/free/en/papers/alva-2025-ai-driven-generation-old-english/) - [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/16/) ## References 7 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer. 2002. Smote: Synthetic minority oversampling technique. Journal of Artificial Intelligence Research, 16:321–357. [find](https://scholar.google.com/scholar?q=N%2E+V%2E+Chawla%2C+K%2E+W%2E+Bowyer%2C+L%2E+O%2E+Hall%2C+and+W%2E+P%2E+Kegelmeyer%2E+2002%2E+Smote%3A+Synthetic+minority+oversampling+technique%2E+Journal+of+Artificial+Intelligence+Research%2C+16%3A321%E2%80%93357%2E) 2. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. [Bert: Pre-training of deep bidirectional transformers for language understanding](https://md.research.elamur.ai/free/en/papers/devlin-2019-bert-pre-training-deep-bidirectional/). [↗](https://doi.org/10.18653/v1/n19-1423) 3. Robbie Love, Claire Dembry, Andrew Hardie, Vaclav Brezina, and Tony McEnery. 2017. The spoken bnc2014: Designing and building a spoken corpus of everyday conversations. International Journal of *Figure 6: Confusion Matrix for Linear SVM Model with Text Features* Corpus Linguistics, 22(3):319–344. © John Benjamins Publishing Company This is an open access article under a OA CC BY license. [find](https://scholar.google.com/scholar?q=Robbie+Love%2C+Claire+Dembry%2C+Andrew+Hardie%2C+Vaclav+Brezina%2C+and+Tony+McEnery%2E+2017%2E+The+spoken+bnc2014%3A+Designing+and+building+a+spoken+corpus+of+everyday+conversations%2E+International+Journal+of+%2AFigur) 4. J. B. MacQueen. 1967. Some methods for classification and analysis of multivariate observations. In Proc. of the fifth Berkeley Symposium on Mathematical Statistics and Probability, volume 1, pages 281–297. University of California Press. [find](https://scholar.google.com/scholar?q=J%2E+B%2E+MacQueen%2E+1967%2E+Some+methods+for+classification+and+analysis+of+multivariate+observations%2E+In+Proc%2E+of+the+fifth+Berkeley+Symposium+on+Mathematical+Statistics+and+Probability%2C+volume+1%2C+pages+28) 5. Leland McInnes, John Healy, and James Melville. 2020. [Umap: Uniform manifold approximation and projection for dimension reduction](https://md.research.elamur.ai/free/en/papers/mcinnes-2018-umap-uniform-manifold-approximation-projection/). 6. Fabian Pedregosa, Gaël Varoquaux, Alexandre Gramfort, Vincent Michel, Bertrand Thirion, Olivier Grisel, Mathieu Blondel, Peter Prettenhofer, Ron Weiss, Vincent Dubourg, et al. 2011. Scikit-learn: Machine learning in python. Journal of machine learning research, 12(Oct):2825–2830. *Figure 8: K-Means Clustering of Utterances (k=2)* P. Rousseeuw. 1987. Silhouette: a graphical aid to the interpretation and validation of cluster analysis. Journal of Computational and Applied Mathematics, 20:53–65. [↗](https://doi.org/10.48550/arxiv.1201.0490) 7. Laurens van der Maaten and Geoffrey Hinton. 2008. Visualizing data using t-SNE. Journal of Machine Learning Research, 9:2579–2605. *Figure 9: Cosine Similarity Between Age Groups* *Figure 7: Mean Utterance Length by Age Bins and Gender* *Figure 10: Model Performance Comparison Bar Plot by Feature Group* *Figure 11: Top Predictive Features for Different Age Groups on One-vs-Rest Logistic Regression Classifier* [find](https://scholar.google.com/scholar?q=Laurens+van+der+Maaten+and+Geoffrey+Hinton%2E+2008%2E+Visualizing+data+using+t%2DSNE%2E+Journal+of+Machine+Learning+Research%2C+9%3A2579%E2%80%932605%2E+%2AFigure+9%3A+Cosine+Similarity+Between+Age+Groups%2A+%2AFigure+7%3A+Mean+Utte) *Updated 1 November 2026*