arXiv subjects

Peter Christen

Publications and source records attributed to Peter Christen.

At least 19 recordsLinked to original sources

arXiv2026

Designing Ambiguity-Aware Clerical Review: A Stratified Sampling Framework for Record Linkage and Deduplication

Clerical review of candidate record pairs remains the de facto gold standard for evaluating record linkage, but it is resource-intensive and often designed informally. We propose a design-based framework that treats clerical review as finite-population sampling over fine-grained strata defined by match weight, comparison pattern, record-level ambiguity, and demographic group. Match-probability bands are constructed from model-based score deciles. Within bands, strata combine comparison patterns with an ambiguity factor derived from matchability and conditional candidate perplexity. A band-specific margin-of-error profile encodes substantive priorities, such as tighter precision in high-score bands, while a single scaling parameter enforces the overall clerical budget. We evaluate the framework using a labelled dataset deduplicated in Splink, comprising 50,000 records and approximately 478,000 candidate pairs. We compare a baseline design reviewing about 23% of pairs with a budget-constrained design reviewing about 7%. The baseline accurately estimates global and band-specific match rates, while sampled distributions of comparison patterns, gender, and ambiguity broadly track the population. Under the budget design, global error approximately doubles, with the largest band-level errors in middle-score bands where matches, non-matches, and ambiguous cases are intermixed. Accuracy in the highest-score bands and the band-wise ambiguity profile are largely preserved, although representativeness by comparison pattern and gender declines. The framework generalises to other clerical-review objectives and can incorporate gold-standard data as prior information for design and calibration. It makes explicit the negotiable trade-offs between workload, precision, representativeness, and coverage of linkage uncertainty.

stat.ME↗

arXiv2026

Learning from Anonymized and Incomplete Tabular Data

User-driven privacy allows individuals to control whether and at what granularity their data is shared, leading to datasets that mix original, generalized, and missing values within the same records and attributes. While such representations are intuitive for privacy, they pose challenges for machine learning, which typically treats non-original values as new categories or as missing, thereby discarding generalization semantics. For learning from such tabular data, we propose novel data transformation strategies that account for heterogeneous anonymization and evaluate them alongside standard imputation and LLM-based approaches. We employ multiple datasets, privacy configurations, and deployment scenarios, demonstrating that our method reliably regains utility. Our results show that generalized values are preferable to pure suppression, that the best data preparation strategy depends on the scenario, and that consistent data representations are crucial for maintaining downstream utility. Overall, our findings highlight that effective learning is tied to the appropriate handling of anonymized values.

cs.LG↗

arXiv2025

Information Leakage in Data Linkage

The process of linking databases that contain sensitive information about individuals across organisations is an increasingly common requirement in the health and social science research domains, as well as with governments and businesses. To protect personal data, protocols have been developed to limit the leakage of sensitive information. Furthermore, privacy-preserving record linkage (PPRL) techniques have been proposed to conduct linkage on encoded data. While PPRL techniques are now being employed in real-world applications, the focus of PPRL research has been on the technical aspects of linking sensitive data (such as encoding methods and cryptanalysis attacks), but not on organisational challenges when employing such techniques in practice. We analyse what sensitive information can possibly leak, either unintentionally or intentionally, in traditional data linkage as well as PPRL protocols, and what a party that participates in such a protocol can learn from the data it obtains legitimately within the protocol. We also show that PPRL protocols can still result in the unintentional leakage of sensitive information. We provide recommendations to help data custodians and other parties involved in a data linkage project to identify and prevent vulnerabilities and make their project more secure.

cs.CR↗

arXiv2024

Efficient Model Repository for Entity Resolution: Construction, Search, and Integration

Entity resolution (ER) is a fundamental task in data integration that enables insights from heterogeneous data sources. The primary challenge of ER lies in classifying record pairs as matches or nonmatches, which in multi-source ER (MS-ER) scenarios can become complicated due to data source heterogeneity and scalability issues. Existing methods for MS-ER generally require labeled record pairs, and such methods fail to effectively reuse models across multiple ER tasks. We propose MoRER (Model Repositories for Entity Resolution), a novel method for building a model repository consisting of classification models that solve ER problems. By leveraging feature distribution analysis, MoRER clusters similar ER tasks, thereby enabling the effective initialization of a model repository with a moderate labeling effort. Experimental results on three multi-source datasets demonstrate that MoRER achieves comparable or better results to methods that have label-limited budgets, such as active learning and transfer learning approaches, while outperforming self-supervised approaches that utilize large pre-trained language models. When compared to supervised transformer-based methods, MoRER achieves comparable or better results, depending on the size of the training data set used.

cs.DB↗

arXiv2024

Anomaly Detection and Classification in Knowledge Graphs

Anomalies such as redundant, inconsistent, contradictory, and deficient values in a Knowledge Graph (KG) are unavoidable, as these graphs are often curated manually, or extracted using machine learning and natural language processing techniques. Therefore, anomaly detection is a task that can enhance the quality of KGs. In this paper, we propose SEKA (SEeking Knowledge graph Anomalies), an unsupervised approach for the detection of abnormal triples and entities in KGs. SEKA can help improve the correctness of a KG whilst retaining its coverage. We propose an adaption of the Path Rank Algorithm (PRA), named the Corroborative Path Rank Algorithm (CPRA), which is an efficient adaptation of PRA that is customized to detect anomalies in KGs. Furthermore, we also present TAXO (TAXOnomy of anomaly types in KGs), a taxonomy of possible anomaly types that can occur in a KG. This taxonomy provides a classification of the anomalies discovered by SEKA with an extensive discussion of possible data quality issues in a KG. We evaluate both approaches using the four real-world KGs YAGO-1, KBpedia, Wikidata, and DSKG to demonstrate the ability of SEKA and TAXO to outperform the baselines.

cs.LG↗

arXiv2024

Selecting a classification performance measure: matching the measure to the problem

The problem of identifying to which of a given set of classes objects belong is ubiquitous, occurring in many research domains and application areas, including medical diagnosis, financial decision making, online commerce, and national security. But such assignments are rarely completely perfect, and classification errors occur. This means it is necessary to compare classification methods and algorithms to decide which is ``best'' for any particular problem. However, just as there are many different classification methods, so there are many different ways of measuring their performance. It is thus vital to choose a measure of performance which matches the aims of the research or application. This paper is a contribution to the growing literature on the relative merits of different performance measures. Its particular focus is the critical importance of matching the properties of the measure to the aims for which the classification is being made.

cs.LG↗

arXiv2023

A Critical Re-evaluation of Benchmark Datasets for (Deep) Learning-Based Matching Algorithms

Entity resolution (ER) is the process of identifying records that refer to the same entities within one or across multiple databases. Numerous techniques have been developed to tackle ER challenges over the years, with recent emphasis placed on machine and deep learning methods for the matching phase. However, the quality of the benchmark datasets typically used in the experimental evaluations of learning-based matching algorithms has not been examined in the literature. To cover this gap, we propose four different approaches to assessing the difficulty and appropriateness of 13 established datasets: two theoretical approaches, which involve new measures of linearity and existing measures of complexity, and two practical approaches: the difference between the best non-linear and linear matchers, as well as the difference between the best learning-based matcher and the perfect oracle. Our analysis demonstrates that most of the popular datasets pose rather easy classification tasks. As a result, they are not suitable for properly evaluating learning-based matching algorithms. To address this issue, we propose a new methodology for yielding benchmark datasets. We put it into practice by creating four new matching tasks, and we verify that these new benchmarks are more challenging and therefore more suitable for further advancements in the field.

cs.DB↗

arXiv2022

Privacy in Practice: Private COVID-19 Detection in X-Ray Images (Extended Version)

Machine learning (ML) can help fight pandemics like COVID-19 by enabling rapid screening of large volumes of images. To perform data analysis while maintaining patient privacy, we create ML models that satisfy Differential Privacy (DP). Previous works exploring private COVID-19 models are in part based on small datasets, provide weaker or unclear privacy guarantees, and do not investigate practical privacy. We suggest improvements to address these open gaps. We account for inherent class imbalances and evaluate the utility-privacy trade-off more extensively and over stricter privacy budgets. Our evaluation is supported by empirically estimating practical privacy through black-box Membership Inference Attacks (MIAs). The introduced DP should help limit leakage threats posed by MIAs, and our practical analysis is the first to test this hypothesis on the COVID-19 classification task. Our results indicate that needed privacy levels might differ based on the task-dependent practical threat from MIAs. The results further suggest that with increasing DP guarantees, empirical privacy leakage only improves marginally, and DP therefore appears to have a limited impact on practical MIA defense. Our findings identify possibilities for better utility-privacy trade-offs, and we believe that empirical attack-specific privacy estimation can play a vital role in tuning for practical privacy.

cs.LG↗

arXiv2021

Big Data is not the New Oil: Common Misconceptions about Population Data

Databases covering all individuals of a population are increasingly used for research and decision-making. The massive size of such databases is often mistaken as a guarantee for valid inferences. However, population data have characteristics that make them challenging to use. Various assumptions on population coverage and data quality are commonly made, including how such data were captured and what types of processing have been applied to them. Furthermore, the full potential of population data can often only be unlocked when such data are linked to other databases. Record linkage often implies subtle technical problems, which are easily missed. We discuss a diverse range of misconceptions relevant for anybody capturing, processing, linking, or analysing population data. Remarkably many of these misconceptions are due to the social nature of data collections and are therefore missed by purely technical accounts of data processing. Many of these misconceptions are also not well documented in scientific publications. We conclude with a set of recommendations for using population data.

cs.DB↗

arXiv2021

Large Scale Record Linkage in the Presence of Missing Data

Record linkage is aimed at the accurate and efficient identification of records that represent the same entity within or across disparate databases. It is a fundamental task in data integration and increasingly required for accurate decision making in application domains ranging from health analytics to national security. Traditional record linkage techniques calculate string similarities between quasi-identifying (QID) values, such as the names and addresses of people. Errors, variations, and missing QID values can however lead to low linkage quality because the similarities between records cannot be calculated accurately. To overcome this challenge, we propose a novel technique that can accurately link records even when QID values contain errors or variations, or are missing. We first generate attribute signatures (concatenated QID values) using an Apriori based selection of suitable QID attributes, and then relational signatures that encapsulate relationship information between records. Combined, these signatures can uniquely identify individual records and facilitate fast and high quality linking of very large databases through accurate similarity calculations between records. We evaluate the linkage quality and scalability of our approach using large real-world databases, showing that it can achieve high linkage quality even when the databases being linked contain substantial amounts of missing values and errors.

cs.DB↗

arXiv2021

Accurate and Efficient Suffix Tree Based Privacy-Preserving String Matching

The task of calculating similarities between strings held by different organizations without revealing these strings is an increasingly important problem in areas such as health informatics, national censuses, genomics, and fraud detection. Most existing privacy-preserving string comparison functions are either based on comparing sets of encoded character q-grams, allow only exact matching of encrypted strings, or they are aimed at long genomic sequences that have a small alphabet. The set-based privacy-preserving similarity functions commonly used to compare name and address strings in the context of privacy-preserving record linkage do not take the positions of sub-strings into account. As a result, two very different strings can potentially be considered as an exact match leading to wrongly linked records. Existing set-based techniques also cannot identify the length of the longest common sub-string across two strings. In this paper we propose a novel approach for accurate and efficient privacy-preserving string matching based on suffix trees that are encoded using chained hashing. We incorporate a hashing based encoding technique upon the encoded suffixes to improve privacy against frequency attacks such as those exploiting Benford's law. Our approach allows various operations to be performed without the strings to be compared being revealed: the length of the longest common sub-string, do two strings have the same beginning, middle or end, and the longest common sub-string similarity between two strings. These functions allow a more accurate comparison of, for example, bank account, credit card, or telephone numbers, which cannot be compared appropriately with existing privacy-preserving string matching techniques. Our evaluation on several data sets with different types of strings validates the privacy and accuracy of our proposed approach.

cs.DB↗

arXiv2020

F*: An Interpretable Transformation of the F-measure

The F-measure, also known as the F1-score, is widely used to assess the performance of classification algorithms. However, some researchers find it lacking in intuitive interpretation, questioning the appropriateness of combining two aspects of performance as conceptually distinct as precision and recall, and also questioning whether the harmonic mean is the best way to combine them. To ease this concern, we describe a simple transformation of the F-measure, which we call F* (F-star), which has an immediate practical interpretation.

cs.LG↗

arXiv2020

Pattern Masking for Dictionary Matching

In the Pattern Masking for Dictionary Matching (PMDM) problem, we are given a dictionary $\mathcal{D}$ of $d$ strings, each of length $\ell$, a query string $q$ of length $\ell$, and a positive integer $z$, and we are asked to compute a smallest set $K\subseteq\{1,\ldots,\ell\}$, so that if $q[i]$, for all $i\in K$, is replaced by a wildcard, then $q$ matches at least $z$ strings from $\mathcal{D}$. The PMDM problem lies at the heart of two important applications featured in large-scale real-world systems: record linkage of databases that contain sensitive information, and query term dropping. In both applications, solving PMDM allows for providing data utility guarantees as opposed to existing approaches. We first show, through a reduction from the well-known $k$-Clique problem, that a decision version of the PMDM problem is NP-complete, even for strings over a binary alphabet. We present a data structure for PMDM that answers queries over $\mathcal{D}$ in time $\mathcal{O}(2^{\ell/2}(2^{\ell/2}+\tau)\ell)$ and requires space $\mathcal{O}(2^{\ell}d^2/\tau^2+2^{\ell/2}d)$, for any parameter $\tau\in[1,d]$. We also approach the problem from a more practical perspective. We show an $\mathcal{O}((d\ell)^{k/3}+d\ell)$-time and $\mathcal{O}(d\ell)$-space algorithm for PMDM if $k=|K|=\mathcal{O}(1)$. We generalize our exact algorithm to mask multiple query strings simultaneously. We complement our results by showing a two-way polynomial-time reduction between PMDM and the Minimum Union problem [Chlamt\'{a}\v{c} et al., SODA 2017]. This gives a polynomial-time $\mathcal{O}(d^{1/4+\epsilon})$-approximation algorithm for PMDM, which is tight under plausible complexity conjectures.

cs.DS↗

arXiv2019

Incremental Clustering Techniques for Multi-Party Privacy-Preserving Record Linkage

Privacy-Preserving Record Linkage (PPRL) supports the integration of sensitive information from multiple datasets, in particular the privacy-preserving matching of records referring to the same entity. PPRL has gained much attention in many application areas, with the most prominent ones in the healthcare domain. PPRL techniques tackle this problem by conducting linkage on masked (encoded) values. Employing PPRL on records from multiple (more than two) parties/sources (multi-party PPRL, MP-PPRL) is an increasingly important but challenging problem that so far has not been sufficiently solved. Existing MP-PPRL approaches are limited to finding only those entities that are present in all parties thereby missing entities that match only in a subset of parties. Furthermore, previous MP-PPRL approaches face substantial scalability limitations due to the need of a large number of comparisons between masked records. We thus propose and evaluate new MP-PPRL approaches that find matches in any subset of parties and still scale to many parties. Our approaches maintain all matches within clusters, where these clusters are incrementally extended or refined by considering records from one party after the other. An empirical evaluation using multiple real datasets ranging from 3 to 26 parties each containing up to $5$ million records validates that our protocols are efficient, and significantly outperform existing MP-PPRL approaches in terms of linkage quality and scalability.

cs.DB↗

arXiv2018

Temporal graph-based clustering for historical record linkage

Research in the social sciences is increasingly based on large and complex data collections, where individual data sets from different domains are linked and integrated to allow advanced analytics. A popular type of data used in such a context are historical censuses, as well as birth, death, and marriage certificates. Individually, such data sets however limit the types of studies that can be conducted. Specifically, it is impossible to track individuals, families, or households over time. Once such data sets are linked and family trees spanning several decades are available it is possible to, for example, investigate how education, health, mobility, employment, and social status influence each other and the lives of people over two or even more generations. A major challenge is however the accurate linkage of historical data sets which is due to data quality and commonly also the lack of ground truth data being available. Unsupervised techniques need to be employed, which can be based on similarity graphs generated by comparing individual records. In this paper we present initial results from clustering birth records from Scotland where we aim to identify all births of the same mother and group siblings into clusters. We extend an existing clustering technique for record linkage by incorporating temporal constraints that must hold between births by the same mother, and propose a novel greedy temporal clustering technique. Experimental results show improvements over non-temporary approaches, however further work is needed to obtain links of high quality.

cs.DB↗

arXiv2018

Developing a Temporal Bibliographic Data Set for Entity Resolution

Entity resolution is the process of identifying groups of records within or across data sets where each group represents a real-world entity. Novel techniques that consider temporal features to improve the quality of entity resolution have recently attracted significant attention. However, there are currently no large data sets available that contain both temporal information as well as ground truth information to evaluate the quality of temporal entity resolution approaches. In this paper, we describe the preparation of a temporal data set based on author profiles extracted from the Digital Bibliography and Library Project (DBLP). We completed missing links between publications and author profiles in the DBLP data set using the DBLP public API. We then used the Microsoft Academic Graph (MAG) to link temporal affiliation information for DBLP authors. We selected around 80K (1%) of author profiles that cover 2 million (50%) publications using information in DBLP such as alternative author names and personal web profile to improve the reliability of the resulting ground truth, while at the same time keeping the data set challenging for temporal entity resolution research.

cs.DB↗

arXiv2018

A Decision Tree Approach to Predicting Recidivism in Domestic Violence

Domestic violence (DV) is a global social and public health issue that is highly gendered. Being able to accurately predict DV recidivism, i.e., re-offending of a previously convicted offender, can speed up and improve risk assessment procedures for police and front-line agencies, better protect victims of DV, and potentially prevent future re-occurrences of DV. Previous work in DV recidivism has employed different classification techniques, including decision tree (DT) induction and logistic regression, where the main focus was on achieving high prediction accuracy. As a result, even the diagrams of trained DTs were often too difficult to interpret due to their size and complexity, making decision-making challenging. Given there is often a trade-off between model accuracy and interpretability, in this work our aim is to employ DT induction to obtain both interpretable trees as well as high prediction accuracy. Specifically, we implement and evaluate different approaches to deal with class imbalance as well as feature selection. Compared to previous work in DV recidivism prediction that employed logistic regression, our approach can achieve comparable area under the ROC curve results by using only 3 of 11 available features and generating understandable decision trees that contain only 4 leaf nodes.

cs.LG↗

arXiv2018

Scalable Entity Resolution Using Probabilistic Signatures on Parallel Databases

Accurate and efficient entity resolution is an open challenge of particular relevance to intelligence organisations that collect large datasets from disparate sources with differing levels of quality and standard. Starting from a first-principles formulation of entity resolution, this paper presents a novel Entity Resolution algorithm that introduces a data-driven blocking and record-linkage technique based on the probabilistic identification of entity signatures in data. The scalability and accuracy of the proposed algorithm are evaluated using benchmark datasets and shown to achieve state-of-the-art results. The proposed algorithm can be implemented simply on modern parallel databases, which allows it to be deployed with relative ease in large industrial applications.

cs.DB↗