After years of research on ontology matching, it is reasonable to consider several questions: is the field of ontology matching still making progress? Is this progress significant enough to pursue further research? If so, what are the particularly promising directions? To answer these questions, we review the state of the art of ontology matching and analyze the results of recent ontology matching evaluations. These results show a measurable improvement in the field, the speed of which is albeit slowing down. We conjecture that significant improvements can be obtained only by addressing important challenges for ontology matching. We present such challenges with insights on how to approach them, thereby aiming to direct research into the most promising tracks and to facilitate the progress of the field.

/pdf/ontology-matching-state-of-the-art-and-future-challenges-486jer3je8.pdf

Ontology Matching: State of the Art and Future Challenges

Voluminous amounts of data have been produced, since the past decade as the miniaturization of Internet of things (IoT) devices increases. However, such data are not useful without analytic power. Numerous big data, IoT, and analytics solutions have enabled people to obtain valuable insight into large data generated by IoT devices. However, these solutions are still in their infancy, and the domain lacks a comprehensive survey. This paper investigates the state-of-the-art research efforts directed toward big IoT data analytics. The relationship between big data analytics and IoT is explained. Moreover, this paper adds value by proposing a new architecture for big IoT data analytics. Furthermore, big IoT data analytic types, methods, and technologies for big data mining are discussed. Numerous notable use cases are also presented. Several opportunities brought by data analytics in IoT paradigm are then discussed. Finally, open research challenges, such as privacy, big data mining, visualization, and integration, are presented as future research directions.

Big IoT Data Analytics: Architecture, Opportunities, and Open Research Challenges

This paper is trying to comment on the various NoSQL (Not only Structured Query Language) systems and to make a comparison (using multiple criteria) between them The NoSQL databases were created as a mean to offer high performance (both in terms of speed and size) and high availability at the price of loosing the ACID (Atomic, Consistent, Isolated, Durable) trait of the traditional databases in exchange with keeping a weaker BASE (Basic Availability, Soft state, Eventual consistency) feature Remains to be seen which of the multiple solutions created since the official appearance of the NoSQL concept (which was defined in 1998 and reintroduced in 2009, around which moment several NoSQL solutions emerged; at the present moment there are known over 120 such solutions) are really delivering on these promises of higher performance (although several of them are already used with very good results)

A comparison between several NoSQL databases with comments and notes

A new class of data storage systems, called NoSQL (Not Only SQL), have emerged to complement traditional database systems, with rejection of general ACID transactions as one common feature. Different platforms, and indeed different primitives within one NoSQL platform, can offer various consistency properties, from Eventual Consistency to single-entity ACID. For the platform provider, weaker consistency should allow better availability, lower latency, and other benefits. This paper investigates what consumers observe of the consistency and performance properties of various offerings. We find that many platforms seem in practice to offer more consistency than they promise; we also find cases where the platform offers consumers a choice between stronger and weaker consistency, but there is no observed benefit from accepting weaker consistency properties.

/pdf/data-consistency-properties-and-the-trade-offs-in-commercial-1x843en9ax.pdf

Data Consistency Properties and the Trade-offs in Commercial Cloud Storages: the Consumers' Perspective

Fuzziness in database management systems

In late May, 2008, a group of database researchers, architects, users and pundits met at the Claremont Resort in Berkeley, California to discuss the state of the research field and its impacts on practice. This was the seventh meeting of this sort in twenty years, and was distinguished by a broad consensus that we are at a turning point in the history of the field, due both to an explosion of data and usage scenarios, and to major shifts in computing hardware and platforms. Given these forces, we are at a time of opportunity for research impact, with an unusually large potential for influential results across computing, the sciences and society. This report details that discussion, and highlights the group's consensus view of new focus areas, including new database engine architectures, declarative programming languages, the interplay of structured and unstructured data, cloud data services, and mobile and virtual worlds. We also report on discussions of the community's growth, including suggestions for changes in community processes to move the research agenda forward, and to enhance impact on a broader audience.

/pdf/the-claremont-report-on-database-research-3hqi31o7n7.pdf

The Claremont report on database research

Database research is expanding, with major efforts in system architecture, new languages, cloud services, mobile and virtual worlds, and interplay between structure and text.

Transformer-based masked language models trained on general corpora, such as BERT and RoBERTa, have shown impressive performance on various downstream tasks. In-creasingly, researchers are “ﬁnetuning” these models to improve performance on domain-speciﬁc tasks. Here, we report a broad study in which we applied 14 transformer-based models to 11 scientiﬁc tasks in order to evaluate how downstream performance is affected by changes along various dimensions (e.g., training data, model size, pretraining time, ﬁnetuning length). In this process, we created the largest and most diverse scientiﬁc language model to date, ScholarBERT, by training a 770M-parameter BERT model on an 221B token scientiﬁc literature dataset spanning many disciplines. Counterintuitively, our evaluation of the 14 BERT-based models (seven versions of ScholarBERT, ﬁve science-speciﬁc large language models from the literature, BERT-Base, and BERT-Large) reveals little difference in performance across the 11 science-focused tasks, despite major differences in model size and training data. We argue that our results establish an upper bound for the performance achievable with BERT-based architectures on tasks from the scientiﬁc domain.

Roger Magoulas

Papers

The Claremont report on database research

The Claremont report on database research

ScholarBERT: Bigger is Not Always Better