• Research softwareclose
• Other research productsclose
• 2018-2022close
• Digital Humanities and Cultural Heritage close

It is a short biographical note for the Digital National Museum in Warsaw for the jubilee of 160 years of existence.

This repository is created for sharing materials (e.g., sample data, trained models, and demo files) for our work. The demo files in the repository allow users to run our models on their own data or on sample data that we provide. The repository includes the following four components: A code demonstration of review text preprocessing. (ReviewPreprocess.zip) The lexicon and a code demonstration of using the lexicon to generate input for the two lexicon-based classification models. (LexiconModels.zip) The trained Doc2Vec model and a code demonstration of obtaining Doc2Vec embeddings using this model. (Doc2VecEmbeddings.zip) Trained base-learner classification models (M2, M3, M4), optimized weights for the ensemble model E2, and the trained ensemble model (E3). We also provide a code demonstration of classifying reviews using our proposed models. (ClassificationModels.zip) The data used for building these models can be requested from the Global Emancipation Network for approved uses established in a data use agreement. This work was funded by the National Science Foundation under award #1936331.

Full Changelog: https://github.com/maehr/the-corpus-as-a-network/commits/v0.1.0-alpha If you use this dataset, please cite it using the metadata from this file.

ASReview Insights is an extension to ASReview LAB that extends the software with tools for plotting and extracting the statistical results of several performance metrics. The extension is especially useful in combination with the simulation functionality of ASReview LAB.

Natural language processing (NLP) for detecting lexical semantic change and linguistic linked open data (LLOD) are two areas of research that have shown promising results in the latest years. However, their potential of being considered together for analysing and representing semantic change from a humanistic perspective needs further study and development. The talk will present an overview of theoretical aspects, NLP techniques and LLOD formalisms intended to this purpose, and will focus on a project developed as a humanities use case within the COST Action “Nexus Linguarum - European network for Web-centred linguistic data science.” The discussion will include preliminary thoughts on the conception of a system that combines dictionary information with corpus evidence, and provides multilingual diachronic ontologies for humanities research.

Research portfolio THE PROBLEM: The collection of items available as uploads (left) are broadly concerned with a class of rhythmic events that that contain disturbing elements, or “hiccups”. Example: Firstly, secondly, weirdly and thirdly… Hiccups can occur in sequences that are either logical, procedural or both (editorial). They are e.g. relevant in connection with photogravure editions. Featuring elements that are unexplainable/irrelevant in the sequence. 1, 2, X and 3… (See Didi-Huberman for an in depth analysis of the question. Didi-Huberman, Georges. (2008). La ressemblance par contact—Archaeologie et modernité de l’empreinte. Minuit.) I am broadly scouting for ways of taking stock such odd elements to see if it is possible to intercept the weft of the passage from image- to object perception. This is done by asking a question, showing an image, and providing an answer. The images are the ones contained in the slide-show (featuring the Q&A at the end). Together, the sequence of 26 Q&As with an image, feature a matrix of the type that Christopher Alexander called a pattern (with the interception X added here). The argument for making an account and finding uses for hiccups is: if unattended they leave a long tail, a growing shadow discussed here in Goethe’s and Jung’s terms. If hiccups are understood as elements occurring in a sequence/edition for which there is no rule, nor any currently existing learning algorithm, then they will tend to aggregate. If left unaccounted and unattended they will grow on par with the power and multiplication of computers in human exchange (i.e., a long entropic tail). Hence we here have a candidate model to explain how human and environmental relations could escalate to states of exception in a variety of un/related areas. And alternatives of how to deal with them are within reach of research. A solution that eliminates the problem—or, an answer that eliminates the question—contains no knowledge. SYNOPSIS From the exhibition element shown in Gallery ROM61 at KMD, for the seminar Tracing Rhythm, a carousel of slides were slung in a spiral out of the Lineup called La Kahina. Her journey around the world as the wife of a diplomat called K, features a variety of homes: domestic interiors adapted to a variety of local circumstances and the job. On the backdrop of this adaptive enclosure the life and work of her husband: the Norwegian political positions in the global oil and gas-trade. Between them a middle zone dedicated to mundane and cultural events: La Kahina’s home is never quite private, and K’s diplomatic errands are never quite public. The middle zone is fringe-space between the private and public domains. It is compared with Goethe’s studio and Arne Næss cabin at Tvergastein (Hallingskarvet). From this an idea of a proximal space—smaller than the world but wider than the body—is connected to the come-and-go between studio-work and fieldwork. In this setting, three art works are compared in terms of how they differently combine field- and studio-work: William Kentridge, Geir Harald Samuelsen, Dragoş Gheorghiu. It is suggested that the vectorial sum of field- and studio-work are organised according to 3 material tropes (Karen Barad): entanglement, superposition and intra-action. From this it is further anticipated that complex phenomena will be composite in terms of these 3 tropes. Providing a ground work for understanding current critical turns. The two chief conversation partners are here Bruno Latour’s titles: Down to Earth (2018) and After lockdown (2021). The idea of the long-tail model above came from there. The essay (attached) was written in preparation to the seminar Tracing Rhythm to reach a ground zero from where the images could be at the forefront. The result was a hiccup: that is, it produced a phenomenological equivalent—or, the event—of Shannon’s definition of information as entropy. On account of the multiplication of images perceived owing to the many instrumental shifts in the production of a photogravure, the techniques becomes an experimental laboratory to explore image collapse into object perception. The Kahina lineup attempts to transpose the production of this transition into an exhibition space, in making the image to object perception available to the viewer in the form of a debatable proposition. However, the photogravure process also features an artistic proposition (exposition) as an active model of similar types of problem (outlined above): notably types of processes where object perception is indeterminately anticipated and proposed. Calling on a hammerbolt, or featuring a counter-beat. As an entry/exit device in working with diary materials—as is the case in the National Library residency in which the project is currently hatching—photgravure has a similar function as the wind-rose on maps. One that steers towards the hatching of an object. Most probably a book. The hosting PKU project—Matter Gesture and Soul—is one of the unique arenas where academic pursuits and research can combine with workshop premises of the art field. Matter Gesture and Soul (DIKU/KMD)

Diese Sicherung beinhaltet das Github Repository zur DHd-Jahreskonferenz 2022. Es enthält alle publizierten Konferenzbeiträge im TEI-XML- sowie im PDF-Format, alle publizierten Posterbeiträge im PDF-Format sowie die Metadaten zu allen Beiträgen im XML- und CSV-Format.

Bi-directional Encoder Representations from Transformers (BERT) is the state-of-the-art deep learning model for pre-training natural language processing (NLP) tasks such as sentiment analysis. The BERT model dynamically generates word representations according to the context and semantics using its bi-directional and attention mechanism features. The model, although, improves precision on NLP tasks, is compute-intensive and time-consuming to deploy on mobile or smaller platforms. In this thesis, to address this issue, we use knowledge distillation (KD), a "teacher-student" training technique, to compress the model. We use the BERT model as the "teacher" model to transfer knowledge to student models, ``first-generation'' convolution neural networks, and long-short term memory with attention mechanism (LSTM-atten). We conduct various experiments on sentiment analysis benchmark data sets and show that the “student models” through knowledge distillation have better performance with 70% improvement in accuracy, precision, recall, and F1-score compared to models without KD. We also investigate the convergence rate of student models and compare the results to the existing models in the literature. Finally, we show that compared to the full-size BERT model, our RNN series models are 50 times smaller in size and retain approximately 96% performance on benchmark data sets.

This Rdata file contains: 1) Spatial Points Data Frame with the Oppida -main forthills- of the central coast of Spain. SpatialPoints: coords.x1 coords.x2 [1,] 659918 4314393 [2,] 607669 4366858 [3,] 639648 4423302 [4,] 647630 4379768 [5,] 706224 4388658 [6,] 733498 4395537 [7,] 699955 4360457 [8,] 737358 4338792 [9,] 714869 4318454 Coordinate Reference System (CRS) arguments: +proj=utm +zone=30 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs 2) A Spatial Polygons Data Frame called "mar" with the seas surrounding the Iberian Peninsula with a bbox of "bbox": min max x -139371.4 1214289 y 3877956.2 5007075 Coordinate Reference System (CRS) arguments: +proj=utm +zone=30 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs 3) TSM A spatial poinst object with the coordinates of Edeta SpatialPoints: coords.x1 coords.x2 [1,] 706183.6 4388395 Coordinate Reference System (CRS) arguments: +proj=utm +zone=30 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs

This repository contains the data and code for our paper: A. Hermann, P. Gutiérrez, C. Chauvel, R. Maury, C. Liorzou, E. Willie, I. Phillip, R. Forkel, C. Rzymski, S. Bedford (in review). Artefact geochemistry demonstrates long-distance voyaging in the Polynesian Outliers.