• Digital Humanities and Cultural Heritage
• Publications close
• AE close
• MU close

Whether a document is of historical or contemporary significance, typography plays a crucial role in its composition. From the early days of modern printing, typographic techniques have evolved and transformed, resulting in changes to the features of typography. By analyzing these features, we can gain insights into specific time periods, geographical locations, and messages conveyed through typography. Therefore, in this paper, we aim to investigate the feasibility of training a model to classify serif typeswithout knowledge of the font and character. We also investigate how to train a vectorial-based image model able to group together fonts with similar features. Specifically, we compare the use of state-of-theart image classification methods, such as the EfficientNet-B2 and the Vision Transformer Base model with different patch sizes, and the state-of-the-art fine-grained image classification method, TransFG, on the serif classification task. We also evaluate the use of the DeepSVG model to learn to group fonts with similar features. Our investigation reveals that fine-grained image classification methods are better suited for the serif classification tasks and that leveraging the character labels helps to learn more meaningful font similarities. This repository contains: - Paper published in the Journal of data mining and digital humanities:WasimEtAl_Toward_Automatic_Typography_Analysis__Serif_Classification_and_Font_Similarities.pdf - Two datasets: The first for serif classification consisting of 126666 training and 2914 font-independent testing images in raster format. The second dataset for svg based similarity learning consists of 124010 training and 2914 font-independent testing images. The images have been categorized into sans-serif, linear-serif, slab-serif, and triangular serif by a designer at the EPFL+ECAL Lab font_serif_dataset.zipfont_svg_dataset.zip - model weights:serif_B_16_long_checkpoint.bin: refers to TransFG ViT B/16 Backendserif_B_32_long_checkpoint.bin: refers to TransFG ViT B/32 Backend ViT-B_16.npz ViT-B_32.npz are publicly available at: https://github.com/TACJu/TransFGand were used for training TransFG ViT B/16 Backend and TransFG ViT B/32 Backend - code:https://github.com/TalalWasim/GEST-Serif

Whether a document is of historical or contemporary significance, typography plays a crucial role in its composition. From the earliest forms of writing in Mesopotamia to the early days of modern printing, typographic techniques have evolved and transformed, resulting in changes to the features of typography. By analyzing these features, we can gain insights into specific time periods, geographical locations, and messages conveyed through typography. Therefore, in this paper, we aim to investigate the feasibility of training a model to classify serif types without knowledge of the font and character. We also investigate how to train a vectorial-based image model able to group together fonts with similar features. Specifically, we compare the use of state-of-the-art image classification methods, such as the EfficientNet-B2 and the Vision Transformer Base model with different patch sizes, and the state-of-the-art fine-grained image classification method, TransFG, on the serif classification task. We also evaluate the use of the DeepSVG model to learn to group fonts with similar features. Our investigation reveals that fine-grained image classification methods are better suited for the serif classification tasks and that leveraging the character labels helps to learn more meaningful font similarities. This repository contains: - Paper published in the Journal of data mining and digital humanities: WasimEtAl_Toward_Automatic_Typography_Analysis__Serif_Classification_and_Font_Similarities.pdf - Two datasets: The first for serif classification consisting of 126666 training and 2914 font-independent testing images in raster format. The second dataset for svg based similarity learning consists of 124010 training and 2914 font-independent testing images. The images have been categorized into sans-serif, linear-serif, slab-serif, and triangular serif by a designer at the EPFL+ECAL Lab font_serif_dataset.zip font_svg_dataset.zip - model weights: serif_B_16_long_checkpoint.bin: refers to TransFG ViT B/16 Backend serif_B_32_long_checkpoint.bin: refers to TransFG ViT B/32 Backend ViT-B_16.npz ViT-B_32.npz are publicly available at: https://github.com/TACJu/TransFG and were used for training TransFG ViT B/16 Backend and TransFG ViT B/32 Backend - code: https://github.com/TalalWasim/GEST-Serif

AbstractThere has still been too little detailed work on the protectionism that emerged in the wake of the Great Depression. In this paper we explore the experiences of two countries that have been largely neglected in the literature, the Netherlands and Netherlands East Indies (NEI). How did these traditionally free‐trading economies respond to the Depression? We construct a detailed product‐level database of tariff and non‐tariff barriers to trade on the basis of primary sources. While ad valorem tariff increases in the Netherlands were largely due to deflation, the country protected agriculture and textiles in a number of ways. Once quotas are taken into account, trade restrictiveness indices suggest that protection in the Netherlands and NEI was comparable to protection in the UK and India, respectively. The NEI quota system was largely geared to protecting Dutch exporters, and succeeded in doing so, but the reverse was not true.

Abstract Four intelligent methods of Gene Expression Programming (GEP), Support Vector Machines (SVM), Multivariate Adaptive Regression Splines (MARS), and Random Forests (RF) were employed and run to estimate the reference evapotranspiration (ET0). Based on the four categories of radiation, mass transmission, temperature, and combination models, weather variables are used in four groups as inputs of the intelligent models. To involve all data in the training and testing process of the models, the K-fold cross-validation technique was employed. Finally, evaluation of the presented intelligent models was performed with measured data of the six sets of lysimetric data (two grass species × three soil textures), also the effect of soil texture and species on intelligent models was investigated. The results obtained indicate that the RF models performed the best among the four groups, which is followed by SVM, MARS, and GEP models, respectively. The best performance belonged to the RF model from the Combination group (RF1) with the values of coefficient of determination, R2 = 0.833; mean bias error, MBE = -0.007 mm/day; mean absolute error, MAE = 1.07 mm/day; root mean square error, RMSE = 1.357 mm/day, and scatter index, SI = 0.184. Based on the ranks of accuracy, the order of the models, from the best to the worst, were: RF2, MARS1, SVM1, RF3, MARS2, GEP1, SVM2, GEP2, RF4, SVM3, MARS3, SVM4, GEP3, GEP4, and MARS4, respectively. The estimated ET0 values by the models indicated higher accuracy for the lysimeters with Sandy Loam Soil compared to other lysimeters. The performance comparison of the intelligent models for the planted grass species in lysimeters yielded different results, as the accuracy was higher for the Festuca grass in Sandy Loam soils compared to the Lolium grass, in contrary in clay and Silt Loam soils the performance of Lolium grass was better.

Abstract Morphology is one of the most important branches of river engineering science. Examining the changes in the geometry of sections in rivers that feed large dams is of particular importance. In fact, gaining a better understanding of the river's behavior will reduce potential damages to hydraulic structures and public infrastructure. In this regard, in this research, the morphological changes of Bazoft River, one of the sub-basins of North Karun, which enters the Karun 4 Dam at the end, have been investigated. In this study, using 26 Landsat + ETM and OLI satellite images, the displacement pattern of the main river channel, along with the changes in width and sinuosity resulting from sedimentation and erosion processes in long and short-term intervals leading to the year 2015 and caused by large floods in this period of time was investigated. The results showed that the average displacement of the main channel of the river leading to erosion (sedimentation) for the right and left banks is 14 (15) and 14 (14) percent of the average width of the river, respectively. Also, the net area change rate due to the displacement of the main channel of the river for the right and left banks was obtained as -1.8 and 0.3 hectares per year, respectively. By examining the effect of floods, it was found that the amount of erosion is higher than sedimentation on both coasts, and the left coast is more dynamic than the right coast. The average width of the river in the 30-year period for the four investigated periods is 33.8, 44.7, 40.2, and 38.6 meters, respectively. Also, the width of the Bazoft River did not change much before the biggest flood in 2009 and its value did not exceed 40 meters. But after the flood, the width of the river suddenly increased and these changes seem obvious. The maximum increase in width is related to the second span of the river and its value is equal to 70 meters. Investigations related to four longitudinal intervals along the river's path showed that the sinuosity value in all years in the first interval had an almost constant trend, but in the second and third intervals, its value is higher, so these two intervals are more active than the other two intervals be in the fourth span, which leads to the Karun Dam 4, changes in sinuosity have a downward trend. Also, the results showed that the dewatering of the Karun 4 dam had a significant effect on all the morphological parameters of the Bazoft River in the fourth period. The results of this study perfectly indicated the capability of revealing the river form changes using Satellite Images Time Series (SITS). This homogeneous result indirectly shows that the erosion rate of the upstream basin and the sediment load of the river increased during this period. Therefore, while updating the status of rivers, the effects of protective and structural measures in such rivers can be evaluated in this way and corrected for improvement.

AbstractAbstracts in academic research serve important functions. An abstract is a summary of an entire article or thesis that defines its objectives, research problem, methodology, major findings, and implications. Abstracts help potential readers to decide whether the article is relevant to their work. Thus, it saves their time and effort. This study aims to answer the following questions: (1) To what extent are the move structures in English and Arabic medical abstracts similar or different, (2) How are metadiscourse features similar or different in English and Arabic medical abstracts? The sample of the study consisted of 120 abstracts (60 English and 60 Arabic) extracted from three major medical journals that require bilingual Arabic and English abstracts. The move structure and metadiscourse features in bilingual abstracts were analyzed using Hyland’s models of 2000 and 2005. The move stability was classified according to Kanoksilapatham’s (2005) model. Both Arabic and English abstracts showed similar move structure patterns, with the method and product moves being obligatory; the purpose and conclusion moves were conventional, and the introduction move was optional. English and Arabic abstracts employed interactive metadiscourse features more than interactional ones. The interactive metadiscourse features in Arabic displayed a higher frequency than those in English, but the interactional metadiscourse features had a lower frequency in the Arabic abstracts than those in English. Transitions were the most used form of interactive metadiscourse features, whereas self-mention and hedges were most common among the interactional metadiscourse features in both Arabic and English abstracts. Arabic and English medical abstracts showed a homogenous rhetorical move structure across both languages. However, their metadiscourse features showed a significant divergence. The findings of such a paper may have theoretical and practical implications. They may contribute to contrastive textology and genre studies since they highlight the similarities and differences between English and Arabic article abstracts. The findings can also be used in training researchers in writing research abstracts.