The DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index.
Similarity encoding for learning with dirty categorical variables
Open accessAbstract
For statistical learning, categorical variables in a table are usually considered as discrete entities and encoded separately to feature vectors, e.g., with one-hot encoding. “Dirty” non-curated data give rise to categorical variables with a very high cardinality but redundancy: several categories reflect the same entity. In databases, this issue is typically solved with a deduplication step. We show that a simple approach that exposes the redundancy to the learning algorithm brings significant gains. We study a generalization of one-hot encoding, similarity encoding, that builds feature vectors from similarities across categories. We perform a thorough empirical validation on non-curated tables, a problem seldom studied in machine learning. Results on seven real-world datasets show that similarity encoding brings significant gains in predictive performance in comparison with known encoding methods for categories or strings, notably one-hot encoding and bag of character n-grams. We draw practical recommendations for encoding dirty categories: 3-gram similarity appears to be a good choice to capture morphological resemblance. For very high-cardinalities, dimensionality reduction significantly reduces the computational cost with little loss in performance: random projections or choosing a subset of prototype categories still outperform classic encoding approaches.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Scikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016
Self-citationEfficient Estimation of Word Representations in Vector Space
Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013) · arXiv · doi:10.48550/arxiv.1301.3781 · Cited by 26,568
Binary Codes capable of currecting deletions, insertions, and reversals
Levenshtein, V. I. (1965) · Soviet physics. Doklady · Cited by 14,804
Enriching Word Vectors with Subword Information
Bojanowski, P., Grave, E., Joulin, A., & Mikolov, T. (2017) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00051 · Cited by 9,645 · cited 2×
“2013) can be seen as a matrix factorization on a matrix of word occurrence in local windows; and fastText (Bojanowski et al. 2016), a state-of-the-art approach for supervised learning on text, is based on a low-rank representation of text.”
BackgroundMultiple regression in behavioral research
Kerlinger, F. N., & Pedhazur, E. J. (1973) · Medical Entomology and Zoology · Cited by 4,919
An introduction to latent semantic analysis
Landauer, T. K., Foltz, P. W., & Laham, D. (1998) · Discourse Processes · doi:10.1080/01638539809545028 · Cited by 4,880 · cited 1×
“Statistical approaches to semantics stem from low-rank data reductions of word occurrences: the original LSA (latent semantic analysis) [23] is a PCA of the word occurrence matrix in documents; word2vec [28] can be seen as a matrix factorization on a matrix of word occurrence in local windows; and…”
MethodApplied Multiple Regression/Correlation Analysis for the Behavioral Sciences
Cohen (2013) · doi:10.4324/9780203774441 · Cited by 3,228
Fellegi, I. P., & Sunter, A. B. (1969) · Journal of the American Statistical Association · doi:10.1080/01621459.1969.10501049 · Cited by 2,475
(2000) · Cited by 1,467
Advances in Record-Linkage Methodology as Applied to Matching the 1985 Census of Tampa, Florida
Jaro, M. A. (1989) · Journal of the American Statistical Association · doi:10.1080/01621459.1989.10478785 · Cited by 1,401
Data preparation for data mining
Pyle, D. (1999) · Cited by 1,368
Duplicate Record Detection: A Survey
Elmagarmid, A. K., Ipeirotis, P. G., & Verykios, V. S. (2007) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2007.250581 · Cited by 1,267
Research design and statistical analysis
(2011) · Choice Reviews Online · doi:10.5860/choice.48-4538 · Cited by 1,202
Development of a Stemming Algorithm
Lovins, J. B. (1968) · Mech. Transl. Comput. Linguistics · Cited by 1,115
The State of Record Linkage and Current Research Problems
Winkler, W. E. (1999) · Cited by 1,015
Feature hashing for large scale multitask learning
Weinberger, K., Dasgupta, A., Langford, J., Smola, A., & Attenberg, J. (2009) · Proceedings of the 26th Annual International Conference on Machine Learning · doi:10.1145/1553374.1553516 · Cited by 906
A Survey of Text Similarity Approaches
H.Gomaa, W., & A. Fahmy, A. (2013) · International Journal of Computer Applications · doi:10.5120/11638-7118 · Cited by 823
Interactive deduplication using active learning
Sarawagi, S., & Bhamidipaty, A. (2002) · Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/775047.775087 · Cited by 692
A Survey of Indexing Techniques for Scalable Record Linkage and Deduplication
Christen, P. (2012) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2011.127 · Cited by 679
A Comparison of String Metrics for Matching Names and Records
Fienberg, W. W. C. A. P. R. A. S. (2003) · Cited by 479
Enriching Word Vectors with Subword Information
Bojanowski, P., Grave, É., Joulin, A., & Mikolov, T. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1607.04606 · Cited by 384
Cohen, W. W. (1998) · Proceedings of the 1998 ACM SIGMOD international conference on Management of data · doi:10.1145/276304.276323 · Cited by 379
Stemming algorithms: A case study for detailed evaluation
Hull, D. A. (1996) · Journal of the American Society for Information Science · doi:10.1002/(sici)1097-4571(199601)47:1<70::aid-asi7>3.3.co;2-q · Cited by 369
Entity Embeddings of Categorical Variables
Guo, C., & Berkhahn, F. (2016) · arXiv · doi:10.48550/arxiv.1604.06737 · Cited by 308 · cited 3×
“Some studies have considered encoding high-cardinality categorical variables [27,16], but not the problem of dirty data.”
BackgroundMethodKim, W., Choi, B.-J., Hong, E.-K., Kim, S.-K., & Lee, D. (2003) · Data Mining and Knowledge Discovery · doi:10.1023/a:1021564703268 · Cited by 301
Micci-Barreca, D. (2001) · ACM SIGKDD Explorations Newsletter · doi:10.1145/507533.507538 · Cited by 276
ActiveClean: interactive data cleaning for statistical modeling
Krishnan, S., Wang, J., Wu, E., Franklin, M. J., & Goldberg, K. (2016) · Proceedings of the VLDB Endowment · doi:10.14778/2994509.2994514 · Cited by 264
Automatic spelling correction using a trigram similarity measure
Angell, R. C., Freund, G. E., & Willett, P. (1983) · Information Processing & Management · doi:10.1016/0306-4573(83)90022-5 · Cited by 175
Davis, M. J. (2021) · Journal of Data Science · doi:10.6339/jds.2010.08(1).563 · Cited by 173
Categorical Variables in Regression Analysis: A Comparison of Dummy and Effect Coding
Alkharusi, H. (2012) · International Journal of Education · doi:10.5296/ije.v4i2.1962 · Cited by 167
Partitioning Nominal Attributes in Decision Trees
Coppersmith, D., Hong, S. J., & Hosking, J. R. M. (1999) · Data Mining and Knowledge Discovery · doi:10.1023/a:1009869804967 · Cited by 159
BoostClean: Automated Error Detection and Repair for Machine LearningPreprint
Krishnan, S., Franklin, M. J. M., Goldberg, K. Y., & Wu, E. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1711.01299 · Cited by 137
Handling Nominal Features in Anomaly Intrusion Detection Problems
Shyu, M., Sarinnapakorn, K., Kuruppu-Appuhamilage, I., Chen, S., Chang, L., & Goldring, T. (2005) · doi:10.1109/ride.2005.10 · Cited by 62
Cited by
Kraus, M., Feuerriegel, S., & Oztekin, A. (2020) · European Journal of Operational Research · doi:10.1016/j.ejor.2019.09.018 · Cited by 392
Li, G., Iyer, B., Prasath, V. B. S., Ni, Y., & Salomonis, N. (2021) · Briefings in Bioinformatics · doi:10.1093/bib/bbab160 · Cited by 168
Sachan, S., Almaghrabi, F., Yang, J.-B., & Xu, D.-L. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.115597 · Cited by 66
Metadata
- Authors
- Cerda, P., Varoquaux, G., & Kégl, B.
- Venue
- Machine Learning
- Published
- 2018-06-12
- Volume
- 107
- Issue
- 8-10
- Pages
- 1477-1494
- ISSN
- 0885-6125 (ISSN-L)
- 1573-0565
- OpenAlex
- W2805979482
- Cited by
- 267
- References
- 33
- FWCI
- 13.59
- Type
- journal article
- Language
- en
- Open access
- bronze
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.