Small Collection, Enormous Power

The contribution of small collections to species distribution modelling: A case study from Fuireneae (Cyperaceae)

Deep learning-based approaches to Handwritten Text Recognition (HTR) have shown remarkable results on publicly available large datasets, both modern and historical. However, it is often the case that historical manuscripts are preserved in small collections, most of the time with unique characteristics in terms of paper support, author handwriting style, and language. State-of-the-art HTR approaches struggle to obtain good performance on such small manuscript collections, for which few training samples are available. In this paper, we focus on HTR on small historical datasets and propose a new historical dataset, which we call Leopardi, with the typical characteristics of small manuscript collections, consisting of letters by the poet Giacomo Leopardi, and devise strategies to deal with the training data scarcity scenario. In particular, we explore the use of carefully designed but cost-effective synthetic data for pre-training HTR models to be applied to small single-author manuscripts. Extensive experiments validate the suitability of the proposed approach, and both the Leopardi dataset and synthetic data will be available to favor further research in this direction.

The anatomical collection of Frederik Ruijsch (1638–1731) concerning injected dry and wet preparations attracted the attention and interest of several generations of scientists and other admirers, including Tsar Peter the Great of Russia (1672–1725). Even today, this interest has not weakened since its manufacture more than three hundred years ago. The main collection of the famous Dutchman was purchased by Tsar Peter the Great in 1717 for his Kunstkamera, now Peter the Great’s Museum of Anthropology and Ethnography in Saint Petersburg. In books and articles, first written by A. I. Tarenetsky, later by M. A. Tikotin, and again by V. V. Ginzburg was pointed out that several dozens of preparations of Frederik Ruijsch are located in the anatomical museum of the Military Medical Academy named S. M. Kirov in St. Petersburg. This collection is called “the small collection” and was acquired by Peter the Great in 1697–1698 as a private collection. This collection was intended for educational purposes and is nowadays almost unknown in, and outside the scientific world. We set ourselves the task to study the history of the acquisition of this small collection, to find out its journey to the end destination by researching early published or unpublished sources. The small collection paved the way for other botanical, zoological, and anatomical collections and has significant historical importance.

Effective management of natural science collections is crucial to maximizing their scientific value. They often contain invaluable items like type specimens and reference material, and data that contributes significantly to our understanding of biodiversity, ecology, and evolution. Research institutes often harbor several smaller collections, which are seldom managed properly, as collection management is not their core business, and because they are not directly connected to the natural science collection community. Recognizing these challenges, DiSSCo Flanders (Distributed System of Scientific Collections) has set out to develop a standardized infrastructure for managing natural science collections, ensuring their proper long-term conservation, and facilitating future reuse (Trekels 2022). Within the DiSSCo Flanders consortium of 14 Belgian institutes, we inventoried the collections utilizing the SYNTHESYS+ survey (Van Baelen 2022) and evaluated their state and management practices using the Collections Self-Assessment Tool (CSAT) (Van Baelen, in preparation). At the Flanders Research Institute for Agriculture, Fisheries and Food (ILVO, Belgium), 15 collections comprising over 558,000 items were cataloged. A significant portion of these, approximately 95%, are marine fish otoliths used for age determination and stock assessments of commercially important fish species. Additionally, ILVO has around 21,000 food- and plant-related items housed within microbial collections. These are instrumental in research related to plant pathology, food spoilage, and biocontrol methods. Furthermore, ILVO maintains several smaller, yet equally important collections. These include insect specimens, macrobenthos reference specimens, an array of seeds and azalea cultivars, marine and agricultural soil samples, and DNA samples from various of the aforementioned collections. To align with the FAIR principles (Findable, Accessible, Interoperable, Reusable) and to connect with the DiSSCo Europe Research Infrastructure, ILVO developed an in-house Collection Management System (CMS). This web application, utilizing a normalized relational database structure, generates dynamic and customizable web pages. It allows users to input, edit, search, and export data through a user-friendly, browser-based interface that supports multiple languages. Data can also be uploaded using custom Microsoft® Excel® templates that conform to the specific data model, ensuring that data is consistently formatted and easily integrable. The back end of the system is responsible for managing the collections and vocabularies (largely aligned with the Darwin Core standard (Darwin Core Task Group. 2009)), while the front end dynamically generates the multilingual user interface and collection-specific vocabularies. This flexibility provides a tailored experience for users based on the specific nature of the collections they are managing, which is particularly important for an institute like ILVO, with diverse collections that require different terminologies and data structures. ILVO-CMS also boasts features that extend beyond basic data management. Users can attach documents and pictures; they can create relationships between items; there are functionalities for referencing other databases; and the system is able to track transactions, such as loans. Though custom-developed for ILVO, the platform’s adaptability enables easy customization and implementation for other institutes and (smaller) collections globally. The usage of ILVO-CMS at ILVO will significantly improve the management, organization, accessibility, and exchangeability of the collections, enhancing their scientific value. By implementing such a comprehensive management tool, ILVO is better positioned to ensure that the collections are not only preserved for future generations but also readily available for ongoing and future research projects. One of the key future goals for ILVO-CMS is the incorporation of 3D images. These virtual representations of physical items will enhance the ability to study and share specimens, through their 3D images, without the need for physical handling, thereby reducing the risk of damage or loss. Additionally, we are committed to the further development of a user-friendly interface to facilitate the (re-)use of ILVO collections, both internally and externally. This includes integrating and updating features based on user feedback and linking a taxonomic backbone, such as that of the Global Biodiversity Information Facility (GBIF Secretariat 2023). ILVO’s experience demonstrates that through opportunities like DiSSCo Flanders, which bring in temporary funding and promote a culture of collaboration, research institutions without public-oriented collections may contribute to global natural science collections.

The basis for this report lies largely in a collection of reptiles and amphibians made in Sonora by the authors in the course of three trips to that state. During late January and early February, 1950, the junior author obtained a few turtles incidental to fish collecting. At the same time, the senior author accompanied William G. Reeder south as far as Guirocoba, Sonora. On this latter trip, about 75 lizards and frogs were collected, all in the southern portion of the state. In August, 1950, the authors accompanied by Mr. Reeder, journeyed south to Guirocoba where two weeks were spent. Although some of the 123 amphibians, 123 snakes, 184 lizards and 21 turtles taken on this trip were obtained enroute between Guirocoba and the International Boundary, the vast majority came from Guirocoba. The size of the Guirocoba collection is a tribute to the ability of the native collectors, since climate and bacilli combined to reduce greatly the gringo's efficiency. Bogert and Oliver (1945) have provided a description of Guirocoba. Briefly, it lies in the Tropical Deciduous Forest about 28 miles by road east southeast of Alamos. Dr. L. M. Klauber has kindly made available to us Sonoran reptiles in his own personal collection and in the collection of the San Diego Society of Natural History. These collections have yielded several new records for the state. A small collection from the vicinity of Guaymas was kindly donated to us by A. A. Allanson of Scripps Institution of Oceanography. Another small but valuable collection was presented by Mr. Sterling Bunnell of Berkeley, California.

Natural history collections have existed for considerable time, and their contribution to research has been discussed and praised in recent decades. In scientific literature, however, there is a general lack of records from private and other small collections. Here, we show that these collections represent a highly valuable resource for research, because they may include an important number of specimens with a broad range of origins. We used the Iberian lynx to demonstrate that the wider and less-biased representation of specimens often found in these collections allows for additional and better inferences than those that are drawn exclusively from large institutions. Locating small zoological collections, however, is very time consuming, and, unfortunately, such collections often disappear quickly, putting their long-term persistence at risk. We propose that authorities, researchers, and curators work together to locate and legalize these specimens and facilitate their inclusion in public databases and, eventually, in larger natural history museums that will ensure their existence in perpetuity.

Cluster afforestation for creating diverse mountain forest structures — a review

When a query jointly addresses very large and very small collections it may happen that an iteration caused by a query operator is driven by a large collection and in each cycle it evaluates a subquery that depends on an element of a small collection. For each such element the result returned by the subquery is the same. In effect, such a subquery is unnecessarily evaluated many times. The optimization rewrites such a query to reverse the situation: the loop is to be performed on a small collection and inside each its cycle a subquery addressing a large collection is evaluated. We illustrate the method on comprehensive examples and then present the general rewriting rule. The research follows the Stack-Based Approach to query languages having roots in the semantics of programming languages. The optimization method consists in analyzing of scoping and binding rules for names occurring in queries.

Peter the Great (1672–1725) – the Russian emperor travelled through Europe during the so-called “Great Embassies” to acquire knowledge in the field of science and industrial production. In Amsterdam (Netherlands), he received as a gift from Frederik Ruijsch samples of dried and wet anatomical specimens, consisting of objects of both natural history and human origin, and currently called the “small collection”. The anatomical part of the collection consisted of 11 dry and 13 wet specimens, parts of the human body. The authors rediscovered this collection in 2016 among the exhibits of the fundamental museum of the department of normal anatomy of the Military Medical Academy (St. Petersburg, Russia). In the article we describe the current state of this historical collection, which played an important role in the acquisition of a more extensive collection by Peter the Great from Ruysch for the Kunstkamera (Museum of Anthropology and Ethnography of Peter the Great in St. Petersburg). The Kunstkamera collection is open to the general public, and as a result is well known through publications in literature. On the contrary, the so-called “small collection” was, from the moment of its arrival in Russia in 1698, in the personal possession of Peter the Great. He later divided his small collection for educational purposes between the Aptekerskii Prikaz and the medical school of Nicolaas Bidloo. Around 1800, both parts of the small collection were reunited and transferred to the Medical-Surgical Academy. Today, this collection still serves educational purposes, but is not widely available to the general public. As a result, it remains virtually unknown.

(1873). XXIX.—List of lepidoptera in a small collection sent from Peru by Mr. Whitely, with descriptions of the new species. Annals and Magazine of Natural History: Vol. 12, No. 69, pp. 218-230.

The Internet of the 21st Century will radically transform how we interact with knowledge. The rise of the World Wide Web and the information infrastructure have rapidly developed the technologies of collections for independent communities. In the future, online information will be dominated by small collections. The information infrastructure must similarly be radically different to support indexing of community collections and searching across such small collections. Users will consider themselves to be navigating in the Interspace, across logical spaces of semantic indexes, rather than in the Internet, across physical networks of computer servers. The digital libraries initiative (DLI) project at the University of Illinois at Urbana-Champaign (UIUC) was one of six sponsored by the NSF, DARPA, and NASA from 1994 through 1998. The goal was to develop widely usable Web technology to effectively search technical documents on the Internet. This article details their efforts.

While the approximate number system (ANS) has been shown to represent relations between numerosities starting in infancy, little is known about whether parallel individuation – a system dedicated to representing objects in small collections – can also be used to represent numerical relations between collections. To test this, we asked preschoolers between the ages of 2 ½ and 4 ½ to compare two arrays of figures that either included exclusively small numerosities (< 4) or exclusively large numerosities (> 4). The ratios of the comparisons were the same in both small and large numerosity conditions. Experiment 1 used a between-subject design, with different groups of preschoolers comparing small and large numerosities, and found that small numerosities are easier to compare than large ones. Experiment 2 replicated this finding with a wider range of set sizes. Experiment 3 further replicated the small-large difference in a within-subject design. We also report tentative evidence that non- and 1-knowers perform better on comparing small numerosities than large numerosities. These results suggest that preschoolers can use parallel individuation to compare numerosities, possibly prior to the onset of number word learning, and thus support previous proposals that there are numerical operations defined over parallel individuation (e.g., Feigenson & Carey, 2003; https://doi.org/10.1111/1467-7687.00313).

Mathematically we can say that people become unhappier with the time if they keep their jobs. To solve this problem we will analyze the evolution of wages in relation with the effort level when the employee maintains his job or he is promoted. We will suppose that we have a small collectivity and we will have lifetime employment and we will prove that the promotion model is based on a Markov chain.

Vicki Funk's legacy continues – The IAPT Small Collections Grant

This paper describes a method, using Genetic Programming, to automatically determine term weighting schemes for the vector space model. Based on a set of queries and their human determined relevant documents, weighting schemes are evolved which achieve a high average precision. In Information Retrieval (IR) systems, useful information for term weighting schemes is available from the query, individual documents and the collection as a whole. We evolve term weighting schemes in both local (within-document) and global (collection-wide) domains which interact with each other correctly to achieve a high average precision. These weighting schemes are tested on well-known test collections and are compared to the traditional tf-idf weighting scheme and to the BM25 weighting scheme using standard IR performance metrics. Furthermore, we show that the global weighting schemes evolved on small collections also increase average precision on larger TREC data. These global weighting schemes are shown to adhere to Luhn's resolving power as both high and low frequency terms are assigned low weights. However, the local weightings evolved on small collections do not perform as well on large collections. We conclude that in order to evolve improved local (within-document) weighting schemes it is necessary to evolve these on large collections.