Our Work

The deposit contains seven self-contained MATLAB scripts that generate all figures and tables of the paper and of the Online Supplementary Material. Figure1_CanonicalModel.m produces Figure 1. Section…

The deposit contains seven self-contained MATLAB scripts that generate all figures and tables of the paper and of the Online Supplementary Material. Figure1_CanonicalModel.m produces Figure 1. Section6_Figures.m produces Figures 2, 3, 4, and Figure A1. Section6_Tables.m produces the numerical values underlying Tables 1–5 of the paper and Table SM.1 of the SM. SM_Figures.m produces Figures SM.1–SM.3. SM_Accelerator.m produces Table SM.4 (financial accelerator extension). SM_MultipleEquilibria_Canonical.m produces Figures SM.4–SM.7 and Table SM.2 (canonical model with three hyperbolic equilibria). SM_MultipleEquilibria_HANK.m produces Figure SM.8 and Table SM.3 (HANK model with two hyperbolic equilibria). The deposit also contains a README with reproduction instructions, an MIT license, and a listing of the expected numerical outputs. It does not duplicate the paper's figures and tables, which are available in the article and the SM. The Online Supplementary Material is included as a related file.

Universities play a fundamental role in the production of scientific knowledge and technological innovation. Theses and dissertations frequently present methods, systems, and conceptual models with po…

Universities play a fundamental role in the production of scientific knowledge and technological innovation. Theses and dissertations frequently present methods, systems, and conceptual models with potential for technological applications and intellectual property registration. However, the identification of these innovative elements is still predominantly carried out manually by specialists working in innovation centers or technology transfer offices, a process that is time-consuming and makes it difficult to analyze large volumes of academic production. In this context, this study aims to present the development of an Artificial Intelligence system based on a multi-agent architecture to support the identification of potential innovations in academic works. The research, developed as a postdoctoral project, was conducted at the Referential Center for Intellectual Property and Innovation in Information Science (CERPII), linked to São Paulo State University (UNESP), Marília campus. The methodology involved the application of Natural Language Processing techniques and language models based on the Transformer architecture to perform semantic analysis of theses and dissertations, considering innovation criteria defined by the Brazilian National Institute of Industrial Property (INPI), such as novelty, inventive step, and applicability. Preliminary results, obtained from the analysis of more than 130 academic documents, indicate that the system contributes to the early identification of research with potential for technological registration. It is concluded that the use of Artificial Intelligence can expand the analytical capacity of university innovation centers, strengthening the relationship between scientific production and technological innovation.Conference abstract published in the proceedings of TOI2026 — XII Congresso Internacional em Tecnologia e Organização da Informação. ISSN 2965-4130.Original publication in Doity

Originally published on the website of the Associazione Italiana Acquario Mediterraneo (AIAM). Deposited in 2026 to preserve the record after the association closed.

Package Moved {text2vec} to Suggests because of archival of {float} code_extend_glove() now checks whether {text2vec} is installed Suggested packages now only offered for installation in interactiv…

Package Moved {text2vec} to Suggests because of archival of {float} code_extend_glove() now checks whether {text2vec} is installed Suggested packages now only offered for installation in interactive sessions Examples and tests requiring suggested packages are now skipped if not installed Now depends on R >= 4.1.0 and uses the native pipe |> internally %>% is still re-exported for users Updated GitHub actions and PR template PR checks now include PR metadata and reverse dependency checks Releases now draw their notes from the NEWS file Data Fixed invalid date in emperors$UNRV causing an error in consolidate() eg

Tezisda neft quduqlarining mahsuldorligini oshirishda qo‘llanadigan issiqlik bilan ishlov berish usullari tahlil qilinadi. Issiq neft yoki suv bilan yuvish, bug‘ bilan siklik ishlov, elekt…

Tezisda neft quduqlarining mahsuldorligini oshirishda qo‘llanadigan issiqlik bilan ishlov berish usullari tahlil qilinadi. Issiq neft yoki suv bilan yuvish, bug‘ bilan siklik ishlov, elektr qizdirish hamda quduq tubida issiqlik hosil qiluvchi tizimlarning vazifalari farqlanadi. Usulni tanlashda parafin-smola yotqiziqlarining joylashuvi, neft xossalari, kollektor tuzilishi, quduq konstruksiyasi va issiqlik yo‘qotishlarini hisobga olish zarurligi asoslanadi. Kon sinovlari o‘tkazilmagan; samarani tekshirish uchun ishlab chiqarish ko‘rsatkichlarini taqqoslash tartibi taklif etiladi.

Introduction. The widespread digital transformation of society leads to changes in traditional forms of leisure, replacing them with digital alternatives. These changes have a significant impact …

Introduction. The widespread digital transformation of society leads to changes in traditional forms of leisure, replacing them with digital alternatives. These changes have a significant impact on student youth. The low controllability of such practices necessitates systematic monitoring of their prevalence, implementation, and potential consequences. The purpose of the article is to highlight the key trends in the transformation of the structure of student leisure in the Sverdlovsk Region, in the context of digitalization, based on long-term monitoring data (1995–2024).Materials and Methods. The data were collected during the IX wave (March–June 2024) of the longitudinal monitoring project “Students of the Middle Urals” (implemented since 1995). A representative quota sample (n = 2 000) of third-year students from 12 universities in the Sverdlovsk Region was surveyed using an online questionnaire, stratified by gender and fi eld of study (humanities, socio-economics, engineering and technology, natural sciences). A comparative analysis of the dynamics for nine waves was conducted using descriptive statistics, trend analysis and the χ² criterion (IBM SPSS Statistics 22.0).Results. A significant shift has been observed from collective to individual – predominantly digital – leisure activities. Certain trends in content consumption have been identified, notably a growing preference for text-based or expert-driven content alongside declining engagement with visual entertainment formats. Paradoxically, awareness of forms of digital participation has increased even as direct engagement has sharply declined. There is an emerging need to develop educational programs and youth policies aimed at balancing online and offline activities, minimizing social isolation, and leveraging the potential of functional digital leisure (such as for self-development).Conclusion. The systemic transformation of student leisure refl ects broader societal digitalization and is characterized by “digital autism” (the avoidance of social interaction) and the construction of identity through digital achievements rather than interpersonal connections. A paradoxical rise in formal civic engagement points to a search for a sense of belonging amidst digital alienation – a trend likely reinforced by state policy. These findings will inform the development of educational programs and support measures aimed at harmonizing digital and real-world interactions within the student community.

MUSCLE (MUSic-related Citizen Science Listening Experiments) is an infrastructure for reusable online music experiments. It can be used by researchers from musicology, music cognition, and other audio…

MUSCLE (MUSic-related Citizen Science Listening Experiments) is an infrastructure for reusable online music experiments. It can be used by researchers from musicology, music cognition, and other audio related fields to implement experiments through a graphical user interface and Python. MUSCLE presents questions, along with text, audio or visual stimuli, to participants. The participants' response is saved to a database and can be exported to json for analysis in R or other software.

Project of The Latvian Council of Science (LCS) - LZP-2024/1-0092 project: “Genetic tools for feed efficient and sustainable meat production in Latvian sheep breeds” The aim of the project…

Project of The Latvian Council of Science (LCS) - LZP-2024/1-0092 project: “Genetic tools for feed efficient and sustainable meat production in Latvian sheep breeds” The aim of the project is to determine whether the feed efficiency status of Latvian meat sheep breeds could be predicted using a panel of genetic and molecular markers previously found to be associated with divergent FE status in a training population of lambs when fed the same diet. Novelty: to determine the parameters predicting the most productive result of lamb rearing, we set out to develop the cheapest and most effective method for determining markers of feed efficiency - based on molecular and genetic markers obtained from the blood of live lambs. About the project: The costs associated with lambing (buying or keeping sheep) and preparing or purchasing feed are the two largest components of variable costs in sheep raising. Feed costs are high due to poor grain growing conditions in major producing countries, the use of feed grains in ethanol production, and increased competition for land in crop production compared to urban development. Feed efficiency in growing lambs (i.e., the animal's ability to reach a market or adult body weight (BW) with the least amount of feed intake) is a key factor in the sheep industry. Improving FE reduces production costs. Improving FE by 5% can bring economic benefits that are up to four times higher than a 5% increase in average daily gain (ADG). Traditionally, meat breeding programs have focused on outputs due mainly to the routine availability of phenotypic data on outputs or correlated traits. Currently, no marker has successfully explained enough of the variability in feed efficiency that they were used as part of a routine improvement program, as well as according to our data, no genetic parameters for performance and feed efficiency traits are available for sheep. The physiological determinants of feed efficiency or putative biomarkers used to analyze animal-to-animal variation in live lambs could be used as a cost-effective and rapid tool for genetic selection or management decisions. About the data of the project: LZP-2024/1-0092 project data on lamb samples of the year 2024 and 2025, or the A24-A25 group, which consists of 197 intensively fattened lambs from eight breeds. The dataset includes data on feed efficiency indicators: (1) ratio traits: Feed efficiency (FE), Feed conversion ratio (FCR), Relative growth rate (RGR), Kleiber’s ratio (KR), or (2) regression or residual traits: Residual feed intake (RFI), Residual weight gain (RWG) and Residual intake and body weight gain (RIG). Calculations were made by using formulas from Lima et al., 2017 (https://doi.org/10.1590/S1806-92902017001000005). Based on the requirements of the breeding program of the breeds, every year, the offspring of the sire ram, certified for breeding activity, are selected and analyzed to estimate the sire rams. All lambs were born as twins, triplets or quadruplets from different ewes and health status was assessed before inclusion in the study so that there were at least two lambs per sire ram from the breed. This study was carried out in cooperation with Latvian Sheep Breeders' Association at the ram breeding control station.   The data is the property of the participants of the University of Latvia.

Ushbu tezisda ishlab chiqarish jarayonlarida elektr energiyasi sarfini real vaqt rejimida monitoring qilish, energiya samaradorligi ko‘rsatkichlarini hisoblash, me’yordan og‘ishlarni…

Ushbu tezisda ishlab chiqarish jarayonlarida elektr energiyasi sarfini real vaqt rejimida monitoring qilish, energiya samaradorligi ko‘rsatkichlarini hisoblash, me’yordan og‘ishlarni aniqlash va moslashuvchan yuklamalarni optimallashtirishga mo‘ljallangan ko‘p bosqichli algoritm taklif etiladi. Algoritm smart-meter, tok va kuchlanish sensorlari, PLC/SCADA/MES hamda sanoat IoT shlyuzlaridan keladigan vaqt qatorlarini birlashtiradi; ma’lumotlarni tozalash va sinxronlashdan so‘ng energiya bazaviy chizig‘i, maxsus energiya sarfi (SEC), cho‘qqi quvvat, bekor turish energiyasi va EnPI og‘ishi hisoblanadi. Keyingi bosqichda prognoz modeli kutilayotgan iste’molni baholaydi, anomaliya moduli ortiqcha sarfni belgilaydi va ko‘p mezonli optimallashtirish bloki ishlab chiqarish hajmini pasaytirmagan holda energiya, tarif xarajati hamda cho‘qqi yuklamani minimallashtiruvchi boshqaruv qarorini shakllantiradi. Algoritmning ishlash mantig‘i imitatsion 24 soatlik ishlab chiqarish profili orqali namoyish etildi. Natijalar monitoringni faqat vizual nazoratdan yopiq siklli “o‘lchash–tahlil–qaror–teskari aloqa” tizimiga o‘tkazish energiya menejmenti samaradorligini oshirishini ko‘rsatadi.