Roee Shraga

Alexander Brinkmann, Roee Shraga, Christian Bizer

E-commerce platforms require structured product data in the form of attribute-value pairs to offer features such as faceted product search or attribute-based product comparison. However, vendors often provide unstructured product descriptions, necessitating the extraction of attribute-value pairs from these texts. BERT-based extraction methods require large amounts of task-specific training data and struggle with unseen attribute values. This paper explores using large language models (LLMs) as a more training-data efficient and robust alternative. We propose prompt templates for zero-shot and few-shot scenarios, comparing textual and JSON-based target schema representations. Our experiments show that GPT-4 achieves the highest average F1-score of 85% using detailed attribute descriptions and demonstrations. Llama-3-70B performs nearly as well, offering a competitive open-source alternative. GPT-4 surpasses the best PLM baseline by 5% in F1-score. Fine-tuning GPT-3.5 increases the performance to the level of GPT-4 but reduces the model's ability to generalize to unseen attribute values.

Alexander Brinkmann, Roee Shraga, Christian Bizer

The goal of entity resolution is to identify records in multiple datasets that represent the same real-world entity. However, comparing all records across datasets can be computationally intensive, leading to long runtimes. To reduce these runtimes, entity resolution pipelines are constructed of two parts: a blocker that applies a computationally cheap method to select candidate record pairs, and a matcher that afterwards identifies matching pairs from this set using more expensive methods. This paper presents SC-Block, a blocking method that utilizes supervised contrastive learning for positioning records in the embedding space, and nearest neighbour search for candidate set building. We benchmark SC-Block against eight state-of-the-art blocking methods. In order to relate the training time of SC-Block to the reduction of the overall runtime of the entity resolution pipeline, we combine SC-Block with four matching methods into complete pipelines. For measuring the overall runtime, we determine candidate sets with 99.5% pair completeness and pass them to the matcher. The results show that SC-Block is able to create smaller candidate sets and pipelines with SC-Block execute 1.5 to 2 times faster compared to pipelines with other blockers, without sacrificing F1 score. Blockers are often evaluated using relatively small datasets which might lead to runtime effects resulting from a large vocabulary size being overlooked. In order to measure runtimes in a more challenging setting, we introduce a new benchmark dataset that requires large numbers of product offers to be blocked. On this large-scale benchmark dataset, pipelines utilizing SC-Block and the best-performing matcher execute 8 times faster than pipelines utilizing another blocker with the same matcher reducing the runtime from 2.5 hours to 18 minutes, clearly compensating for the 5 minutes required for training SC-Block.

Alexander Brinkmann, Roee Shraga, Reng Chiz Der et al.

Structured product data in the form of attribute/value pairs is the foundation of many e-commerce applications such as faceted product search, product comparison, and product recommendation. Product offers often only contain textual descriptions of the product attributes in the form of titles or free text. Hence, extracting attribute/value pairs from textual product descriptions is an essential enabler for e-commerce applications. In order to excel, state-of-the-art product information extraction methods require large quantities of task-specific training data. The methods also struggle with generalizing to out-of-distribution attributes and attribute values that were not a part of the training data. Due to being pre-trained on huge amounts of text as well as due to emergent effects resulting from the model size, Large Language Models like ChatGPT have the potential to address both of these shortcomings. This paper explores the potential of ChatGPT for extracting attribute/value pairs from product descriptions. We experiment with different zero-shot and few-shot prompt designs. Our results show that ChatGPT achieves a performance similar to a pre-trained language model but requires much smaller amounts of training data and computation for fine-tuning.

Koyena Pal, Aamod Khatiwada, Roee Shraga et al.

Data management has traditionally relied on synthetic data generators to generate structured benchmarks, like the TPC suite, where we can control important parameters like data size and its distribution precisely. These benchmarks were central to the success and adoption of database management systems. But more and more, data management problems are of a semantic nature. An important example is finding tables that can be unioned. While any two tables with the same cardinality can be unioned, table union search is the problem of finding tables whose union is semantically coherent. Semantic problems cannot be benchmarked using synthetic data. Our current methods for creating benchmarks involve the manual curation and labeling of real data. These methods are not robust or scalable and perhaps more importantly, it is not clear how robust the created benchmarks are. We propose to use generative AI models to create structured data benchmarks for table union search. We present a novel method for using generative models to create tables with specified properties. Using this method, we create a new benchmark containing pairs of tables that are both unionable and non-unionable but related. We thoroughly evaluate recent existing table union search methods over existing benchmarks and our new benchmark. We also present and evaluate a new table search methods based on recent large language models over all benchmarks. We show that the new benchmark is more challenging for all methods than hand-curated benchmarks, specifically, the top-performing method achieves a Mean Average Precision of around 60%, over 30% less than its performance on existing manually created benchmarks. We examine why this is the case and show that the new benchmark permits more detailed analysis of methods, including a study of both false positives and false negatives that were not possible with existing benchmarks.

Avigdor Gal, Roee Shraga

Data integration has been recently challenged by the need to handle large volumes of data, arriving at high velocity from a variety of sources, which demonstrate varying levels of veracity. This challenging setting, often referred to as big data, renders many of the existing techniques, especially those that are human-intensive, obsolete. Big data also produces technological advancements such as Internet of things, cloud computing, and deep learning, and accordingly, provides a new, exciting, and challenging research agenda. Given the availability of data and the improvement of machine learning techniques, this blog discusses the respective roles of humans and machines in achieving cognitive tasks in matching, aiming to determine whether traditional roles of humans and machines are subject to change. Such investigation, we believe, will pave a way to better utilize both human and machine resources in new and innovative manners. We shall discuss two possible modes of change, namely humans out and humans in. Humans out aim at exploring out-of-the-box latent matching reasoning using machine learning algorithms when attempting to overpower human matcher performance. Pursuing out-of-the-box thinking, machine and deep learning can be involved in matching. Humans in explores how to better involve humans in the matching loop by assigning human matchers with a symmetric role to algorithmic matcher in the matching process.

Bar Genossar, Roee Shraga, Avigdor Gal

Entity resolution, a longstanding problem of data cleaning and integration, aims at identifying data records that represent the same real-world entity. Existing approaches treat entity resolution as a universal task, assuming the existence of a single interpretation of a real-world entity and focusing only on finding matched records, separating corresponding from non-corresponding ones, with respect to this single interpretation. However, in real-world scenarios, where entity resolution is part of a more general data project, downstream applications may have varying interpretations of real-world entities relating, for example, to various user needs. In what follows, we introduce the problem of multiple intents entity resolution (MIER), an extension to the universal (single intent) entity resolution task. As a solution, we propose FlexER, utilizing contemporary solutions to universal entity resolution tasks to solve multiple intents entity resolution. FlexER addresses the problem as a multi-label classification problem. It combines intent-based representations of tuple pairs using a multiplex graph representation that serves as an input to a graph neural network (GNN). FlexER learns intent representations and improves the outcome to multiple resolution problems. A large-scale empirical evaluation introduces a new benchmark and, using also two well-known benchmarks, shows that FlexER effectively solves the MIER problem and outperforms the state-of-the-art for a universal entity resolution.

Bar Genossar, Avigdor Gal, Roee Shraga

Entity matching, a core data integration problem, is the task of deciding whether two data tuples refer to the same real-world entity. Recent advances in deep learning methods, using pre-trained language models, were proposed for resolving entity matching. Although demonstrating unprecedented results, these solutions suffer from a major drawback as they require large amounts of labeled data for training, and, as such, are inadequate to be applied to low resource entity matching problems. To overcome the challenge of obtaining sufficient labeled data we offer a new active learning approach, focusing on a selection mechanism that exploits unique properties of entity matching. We argue that a distributed representation of a tuple pair indicates its informativeness when considered among other pairs. This is used consequently in our approach that iteratively utilizes space-aware considerations. Bringing it all together, we treat the low resource entity matching problem as a Battleship game, hunting indicative samples, focusing on positive ones, through awareness of the latent space along with careful planning of next sampling iterations. An extensive experimental analysis shows that the proposed algorithm outperforms state-of-the-art active learning solutions to low resource entity matching, and although using less samples, can be as successful as state-of-the-art fully trained known algorithms.

Roee Shraga

This work offers a novel view on the use of human input as labels, acknowledging that humans may err. We build a behavioral profile for human annotators which is used as a feature representation of the provided input. We show that by utilizing black-box machine learning, we can take into account human behavior and calibrate their input to improve the labeling quality. To support our claims and provide a proof-of-concept, we experiment with three different matching tasks, namely, schema matching, entity matching and text matching. Our empirical evaluation suggests that the method can improve the quality of gathered labels in multiple settings including cross-domain (across different matching tasks).

Nicholas Pulsone, Roee Shraga, Gregory Goren

Entity Matching (EM)--the task of determining whether two data records refer to the same real-world entity--is a core task in data integration. Recent advances in deep learning have set a new standard for EM, particularly through fine-tuning Pretrained Language Models (PLMs) and, more recently, Large Language Models (LLMs). However, fine-tuning typically requires large amounts of labeled data, which are expensive and time-consuming to obtain. In the context of e-commerce matching, labeling scarcity varies widely across domains, raising the question of how to intelligently train accurate domain-specific EM models with limited labeled data. In this work we assume users have only limited amount of labels for a specific target domain but have access to labeled data from other domains. We introduce BEACON, a distribution-aware, budget-aware framework for low-resource EM across domains. BEACON leverages the insight that embedding representations of pairwise candidate matches can guide the effective selection of out-of-domain samples under limited in-domain supervision. We conduct extensive experiments across multiple domain-partitioned datasets derived from established EM benchmarks, demonstrating that BEACON consistently outperforms state-of-the-art methods under different training budgets.

Bar Genossar, Sagi Dalyot, Roee Shraga et al.

Urban environments are continuously mapped and modeled by various data collection platforms, including satellites, unmanned aerial vehicles and street cameras. The growing availability of 3D geospatial data from multiple modalities has introduced new opportunities and challenges for integrating spatial knowledge at scale, particularly in high-impact domains such as urban planning and rapid disaster management. Geospatial entity resolution is the task of identifying matching spatial objects across different datasets, often collected independently under varying conditions. Existing approaches typically rely on spatial proximity, textual metadata, or external identifiers to determine correspondence. While useful, these signals are often unavailable, unreliable, or misaligned, especially in cross-source scenarios. To address these limitations, we shift the focus to the intrinsic geometry of 3D spatial objects and present 3dSAGER (3D Spatial-Aware Geospatial Entity Resolution), an end-to-end pipeline for geospatial entity resolution over 3D objects. 3dSAGER introduces a novel, spatial-reference-independent featurization mechanism that captures intricate geometric characteristics of matching pairs, enabling robust comparison even across datasets with incompatible coordinate systems where traditional spatial methods fail. As a key component of 3dSAGER, we also propose a new lightweight and interpretable blocking method, BKAFI, that leverages a trained model to efficiently generate high-recall candidate sets. We validate 3dSAGER through extensive experiments on real-world urban datasets, demonstrating significant gains in both accuracy and efficiency over strong baselines. Our empirical study further dissects the contributions of each component, providing insights into their impact and the overall design choices.

Roee Shraga, Gil Katz, Yael Badian et al.

Industry 4.0 offers opportunities to combine multiple sensor data sources using IoT technologies for better utilization of raw material in production lines. A common belief that data is readily available (the big data phenomenon), is oftentimes challenged by the need to effectively acquire quality data under severe constraints. In this paper we propose a design methodology, using active learning to enhance learning capabilities, for building a model of production outcome using a constrained amount of raw material training data. The proposed methodology extends existing active learning methods to effectively solve regression-based learning problems and may serve settings where data acquisition requires excessive resources in the physical world. We further suggest a set of qualitative measures to analyze learners performance. The proposed methodology is demonstrated using an actual application in the milk industry, where milk is gathered from multiple small milk farms and brought to a dairy production plant to be processed into cottage cheese.

Artem Frenk, Roee Shraga

Our work introduces SAVeD (Semantically Aware Version Detection), a contrastive learning-based framework for identifying versions of structured datasets without relying on metadata, labels, or integration-based assumptions. SAVeD addresses a common challenge in data science of repeated labor due to a difficulty of similar work or transformations on datasets. SAVeD employs a modified SimCLR pipeline, generating augmented table views through random transformations (e.g., row deletion, encoding perturbations). These views are embedded via a custom transformer encoder and contrasted in latent space to optimize semantic similarity. Our model learns to minimize distances between augmented views of the same dataset and maximize those between unrelated tables. We evaluate performance using validation accuracy and separation, defined respectively as the proportion of correctly classified version/non-version pairs on a hold-out set, and the difference between average similarities of versioned and non-versioned tables (defined by a benchmark, and not provided to the model). Our experiments span five canonical datasets from the Semantic Versioning in Databases Benchmark, and demonstrate substantial gains post-training. SAVeD achieves significantly higher accuracy on completely unseen tables in, and a significant boost in separation scores, confirming its capability to distinguish semantically altered versions. Compared to untrained baselines and prior state-of-the-art dataset-discovery methods like Starmie, our custom encoder achieves competitive or superior results.

Sreeram Marimuthu, Nina Klimenkova, Roee Shraga

Data discovery and table unionability in particular became key tasks in modern Data Science. However, the human perspective for these tasks is still under-explored. Thus, this research investigates the human behavior in determining table unionability within data discovery. We have designed an experimental survey and conducted a comprehensive analysis, in which we assess human decision-making for table unionability. We use the observations from the analysis to develop a machine learning framework to boost the (raw) performance of humans. Furthermore, we perform a preliminary study on how LLM performance is compared to humans indicating that it is typically better to consider a combination of both. We believe that this work lays the foundations for developing future Human-in-the-Loop systems for efficient data discovery.

Roee Shraga, Avigdor Gal

Schema matching is a core task of any data integration process. Being investigated in the fields of databases, AI, Semantic Web and data mining for many years, the main challenge remains the ability to generate quality matches among data concepts (e.g., database attributes). In this work, we examine a novel angle on the behavior of humans as matchers, studying match creation as a process. We analyze the dynamics of common evaluation measures (precision, recall, and f-measure), with respect to this angle and highlight the need for unbiased matching to support this analysis. Unbiased matching, a newly defined concept that describes the common assumption that human decisions represent reliable assessments of schemata correspondences, is, however, not an inherent property of human matchers. In what follows, we design PoWareMatch that makes use of a deep learning mechanism to calibrate and filter human matching decisions adhering the quality of a match, which are then combined with algorithmic matching to generate better match results. We provide an empirical evidence, established based on an experiment with more than 200 human matchers over common benchmarks, that PoWareMatch predicts well the benefit of extending the match with an additional correspondence and generates high quality matches. In addition, PoWareMatch outperforms state-of-the-art matching algorithms.

Diego Calvanese, Avigdor Gal, Davide Lanti et al.

Virtual Knowledge Graphs (VKG) constitute one of the most promising paradigms for integrating and accessing legacy data sources. A critical bottleneck in the integration process involves the definition, validation, and maintenance of mappings that link data sources to a domain ontology. To support the management of mappings throughout their entire lifecycle, we propose a comprehensive catalog of sophisticated mapping patterns that emerge when linking databases to ontologies. To do so, we build on well-established methodologies and patterns studied in data management, data analysis, and conceptual modeling. These are extended and refined through the analysis of concrete VKG benchmarks and real-world use cases, and considering the inherent impedance mismatch between data sources and ontologies. We validate our catalog on the considered VKG scenarios, showing that it covers the vast majority of patterns present therein.

Roee Shraga, Ofra Amir, Avigdor Gal

Matching is a task at the heart of any data integration process, aimed at identifying correspondences among data elements. Matching problems were traditionally solved in a semi-automatic manner, with correspondences being generated by matching algorithms and outcomes subsequently validated by human experts. Human-in-the-loop data integration has been recently challenged by the introduction of big data and recent studies have analyzed obstacles to effective human matching and validation. In this work we characterize human matching experts, those humans whose proposed correspondences can mostly be trusted to be valid. We provide a novel framework for characterizing matching experts that, accompanied with a novel set of features, can be used to identify reliable and valuable human experts. We demonstrate the usefulness of our approach using an extensive empirical evaluation. In particular, we show that our approach can improve matching results by filtering out inexpert matchers.