Refine
Year of publication
- 2022 (1399) (remove)
Document Type
- Article (937)
- Doctoral Thesis (179)
- Postprint (157)
- Part of a Book (38)
- Monograph/Edited Volume (28)
- Working Paper (23)
- Conference Proceeding (13)
- Review (11)
- Part of Periodical (4)
- Bachelor Thesis (2)
Language
- English (1399) (remove)
Keywords
- COVID-19 (16)
- machine learning (15)
- climate change (14)
- exercise (10)
- Germany (8)
- analysis (8)
- gender (8)
- obesity (8)
- social media (8)
- Tolkien (7)
Institute
- Institut für Biochemie und Biologie (206)
- Institut für Physik und Astronomie (189)
- Extern (143)
- Institut für Geowissenschaften (111)
- Institut für Chemie (86)
- Strukturbereich Kognitionswissenschaften (70)
- Institut für Umweltwissenschaften und Geographie (68)
- Fachgruppe Betriebswirtschaftslehre (62)
- Fachgruppe Volkswirtschaftslehre (48)
- Hasso-Plattner-Institut für Digital Engineering GmbH (48)
“Broadcast your gender.”
(2022)
Social media platforms provide a large array of behavioral data relevant to social scientific research. However, key information such as sociodemographic characteristics of agents are often missing. This paper aims to compare four methods of classifying social attributes from text. Specifically, we are interested in estimating the gender of German social media creators. By using the example of a random sample of 200 YouTube channels, we compare several classification methods, namely (1) a survey among university staff, (2) a name dictionary method with the World Gender Name Dictionary as a reference list, (3) an algorithmic approach using the website gender-api.com, and (4) a Multinomial Naïve Bayes (MNB) machine learning technique. These different methods identify gender attributes based on YouTube channel names and descriptions in German but are adaptable to other languages. Our contribution will evaluate the share of identifiable channels, accuracy and meaningfulness of classification, as well as limits and benefits of each approach. We aim to address methodological challenges connected to classifying gender attributes for YouTube channels as well as related to reinforcing stereotypes and ethical implications.
“Broadcast your gender.”
(2022)
Social media platforms provide a large array of behavioral data relevant to social scientific research. However, key information such as sociodemographic characteristics of agents are often missing. This paper aims to compare four methods of classifying social attributes from text. Specifically, we are interested in estimating the gender of German social media creators. By using the example of a random sample of 200 YouTube channels, we compare several classification methods, namely (1) a survey among university staff, (2) a name dictionary method with the World Gender Name Dictionary as a reference list, (3) an algorithmic approach using the website gender-api.com, and (4) a Multinomial Naïve Bayes (MNB) machine learning technique. These different methods identify gender attributes based on YouTube channel names and descriptions in German but are adaptable to other languages. Our contribution will evaluate the share of identifiable channels, accuracy and meaningfulness of classification, as well as limits and benefits of each approach. We aim to address methodological challenges connected to classifying gender attributes for YouTube channels as well as related to reinforcing stereotypes and ethical implications.
The Sino-Japanese War of 1894/95 is usually only briefly mentioned in studies on diplomatic history. Especially the war's impact on Wilhelmine foreign and world policy (Weltpolitik) has been largely neglected. However, the events in East Asia had a profound influence on the political leadership in Berlin. The Wilhelmstrasse's attitude towards the conflict changed rapidly when the course of the war in Northeast Asia made a collapse of the Qing Empire increasingly likely. Afraid of the prospect of being left empty handed in an upcoming scramble for China, German diplomacy got active in early 1895. Driven by a hectic activism which soon should become a dominant feature of Weltpolitik, Berlin concluded an ad-hoc alliance with St. Petersburg and Paris. In April 1895, this unlikely coalition intervened against Tokyo. While the Triple Intervention served primarily Russia's interest to maintain the status quo on the Chinese mainland, Germany aimed at the acquisition of a military and commercial base in Northeast Asia. Driven by public opinion, the naval leadership and the Emperor Wilhelm II., the formerly neutral and reserved German diplomacy changed towards an aggressive and unstable imperialist policy, which ultimately resulted in the acquisition of Qingdao in November 1897.
Background
Coronavirus disease (COVID-19) has a severe impact on all aspects of patient care. Among the numerous biomarkers of potential validity for diagnostic and clinical management of COVID-19 are biomarkers at the interface of iron metabolism and inflammation.
Methods
The follow-up study included 54 hospitalized patients with laboratory-confirmed COVID-19 with a moderate and severe/critical form of the disease. Iron deficiency specific biomarkers such as iron, ferritin, transferrin receptor, hepcidin, and zinc protoporphyrin (ZnPP) as well as relevant markers of inflammation were evaluated twice: in the first five days when the patient was admitted to the hospital and during five to 15 days; and their validity to diagnose iron deficiency was further assessed. The regression and Receiver Operating Characteristics (ROC) analyses were performed to evaluate the prognosis and determine the probability for predicting the severity of the disease in the first five days of COVID-19.
Results
Based on hemoglobin values, anemia was observed in 21 of 54 patients. Of all iron deficiency anemia-related markers, only ZnPP was significantly elevated (P<0.001) in the anemic group. When patients were grouped according to the severity of disease, slight differences in hemoglobin or other anemia-related parameters could be observed. However, the levels of ZnPP were significantly increased in the severely ill group of patients. The ratio of ZnPP to lymphocyte count (ZnPP/L) had a discrimination power stronger than the neutrophil to lymphocyte count ratio (N/L) to determine disease severity. Additionally, only two markers were independently associated with the severity of COVID-19 in logistic regression analysis; D-dimer (OR (5.606)(95% CI 1.019–30.867)) and ZnPP/L ratio (OR (74.313) (95% CI 1.081–5108.103)).
Conclusions
For the first time ZnPP in COVID-19 patients were reported in this study. Among all iron-related markers tested, ZnPP was the only one that was associated with anemia as based on hemoglobin. The increase in ZnPP might indicate that the underlying cause of anemia in COVID-19 patients is not only due to the inflammation but also of nutritional origin. Additionally, the ZnPP/L ratio might be a valid prognostic marker for the severity of COVID-19.
Background
Coronavirus disease (COVID-19) has a severe impact on all aspects of patient care. Among the numerous biomarkers of potential validity for diagnostic and clinical management of COVID-19 are biomarkers at the interface of iron metabolism and inflammation.
Methods
The follow-up study included 54 hospitalized patients with laboratory-confirmed COVID-19 with a moderate and severe/critical form of the disease. Iron deficiency specific biomarkers such as iron, ferritin, transferrin receptor, hepcidin, and zinc protoporphyrin (ZnPP) as well as relevant markers of inflammation were evaluated twice: in the first five days when the patient was admitted to the hospital and during five to 15 days; and their validity to diagnose iron deficiency was further assessed. The regression and Receiver Operating Characteristics (ROC) analyses were performed to evaluate the prognosis and determine the probability for predicting the severity of the disease in the first five days of COVID-19.
Results
Based on hemoglobin values, anemia was observed in 21 of 54 patients. Of all iron deficiency anemia-related markers, only ZnPP was significantly elevated (P<0.001) in the anemic group. When patients were grouped according to the severity of disease, slight differences in hemoglobin or other anemia-related parameters could be observed. However, the levels of ZnPP were significantly increased in the severely ill group of patients. The ratio of ZnPP to lymphocyte count (ZnPP/L) had a discrimination power stronger than the neutrophil to lymphocyte count ratio (N/L) to determine disease severity. Additionally, only two markers were independently associated with the severity of COVID-19 in logistic regression analysis; D-dimer (OR (5.606)(95% CI 1.019–30.867)) and ZnPP/L ratio (OR (74.313) (95% CI 1.081–5108.103)).
Conclusions
For the first time ZnPP in COVID-19 patients were reported in this study. Among all iron-related markers tested, ZnPP was the only one that was associated with anemia as based on hemoglobin. The increase in ZnPP might indicate that the underlying cause of anemia in COVID-19 patients is not only due to the inflammation but also of nutritional origin. Additionally, the ZnPP/L ratio might be a valid prognostic marker for the severity of COVID-19.
We present fully polynomial time approximation schemes for a broad class of Holant problems with complex edge weights, which we call Holant polynomials. We transform these problems into partition functions of abstract combinatorial structures known as polymers in statistical physics. Our method involves establishing zero-free regions for the partition functions of polymer models and using the most significant terms of the cluster expansion to approximate them. Results of our technique include new approximation and sampling algorithms for a diverse class of Holant polynomials in the low-temperature regime (i.e. small external field) and approximation algorithms for general Holant problems with small signature weights. Additionally, we give randomised approximation and sampling algorithms with faster running times for more restrictive classes. Finally, we improve the known zero-free regions for a perfect matching polynomial.
In 1988 the youth-led movement "Schools without racism, schools with courage" was established in Belgium and quickly spread throughout Europe. German schools adopted this movement in 1995. Decades later, racism is not yet a strong developmental science research topic for studies of youth in Germany and Europe. In this commentary we argue that it should be. With increasing hate crimes and harassment, there is also a need to understand how families are socializing young people to be prepared for, cope with, resist, and disrupt racism. This type of ethnic-racial socialization affects important developmental processes-adolescent ethnic-racial identity development and intergroup and institutional understanding and relations-and requires a more prominent place of study in a migration-diverse Germany. Studying these issues in this particular sociohistorical context will also contribute to a more context-specific understanding of youth experiences of racism.
Against the background of the increasingly discussed “Linguistic Saving Hypothesis” (Chen, 2013), I studied whether the targeted use of a present tense (close tense) and a future tense (distant tense) within the same language have an impact on intertemporal decision-making. In a monetarily incentivized laboratory experiment in Germany, I implemented two different treatments on intertemporal choices. The treatments differed in the tense in which I referred to future rewards. My results show that individuals prefer to a greater extent rewards which are associated with a present tense (close tense). This result is in line with my prediction and the first empirical support for the Linguistic Saving Hypothesis within one language. However, this result holds exclusively for males. Females seem to be unaffected by the linguistic manipulation. I discuss my findings in the context of “gender-as-culture” as well as their potential policy-implications.
We have analysed an archival XMM-Newton EPIC observation that serendipitously covered the sky position of a variable X-ray source AX J1714.1-3912, previously suggested to be a Supergiant Fast X-ray Transient (SFXT). During the XMM-Newton observation the source is variable on a timescale of hundred seconds and shows two luminosity states, with a flaring activity followed by unflared emission, with a variability amplitude of a factor of about 50. We have discovered an intense iron emission line with a centroid energy of 6.4 keV in the power law-like spectrum, modified by a large absorption (N-H similar to 10(24) cm(-2)), never observed before from this source. This X-ray spectrum is unusual for an SFXT, but resembles the so-called 'highly obscured sources', high mass X-ray binaries (HMXBs) hosting an evolved B[e] supergiant companion (sgB[e]). This might suggest that AX J1714.1-3912 is a new member of this rare type of HMXBs, which includes IGR J16318-4848 and CI Camelopardalis. Increasing this small population of sources would be remarkable, as they represent an interesting short transition evolutionary stage in the evolution of massive binaries. Nevertheless, AX J1714.1-3912 appears to share X-ray properties of both kinds of HMXBs (SFXT versus sgB[e] HMXB). Therefore, further investigations of the companion star are needed to disentangle the two hypothesis.
X-rays are integral to furthering our knowledge of exoplanetary systems. In this work we discuss the use of X-ray observations to understand star-planet interac- tions, mass-loss rates of an exoplanet’s atmosphere and the study of an exoplanet’s atmospheric components using future X-ray spectroscopy.
The low-mass star GJ 1151 was reported to display variable low-frequency radio emission, which is an indication of coronal star-planet interactions with an unseen exoplanet. In chapter 5 we report the first X-ray detection of GJ 1151’s corona based on XMM-Newton data. Averaged over the observation, we detect the star with a low coronal temperature of 1.6 MK and an X-ray luminosity of LX = 5.5 × 1026 erg/s. This is compatible with the coronal assumptions for a sub-Alfvénic star- planet interaction origin of the observed radio signals from this star.
In chapter 6, we aim to characterise the high-energy environment of known ex- oplanets and estimate their mass-loss rates. This work is based on the soft X-ray instrument on board the Spectrum Roentgen Gamma (SRG) mission, eROSITA, along with archival data from ROSAT, XMM-Newton, and Chandra. We use these four X-ray source catalogues to derive X-ray luminosities of exoplanet host stars in the 0.2-2 keV energy band. A catalogue of the mass-loss rates of 287 exoplan- ets is presented, with 96 of these planets characterised for the first time using new eROSITA detections. Of these first time detections, 14 are of transiting exoplanets that undergo irradiation from their host stars that is of a level known to cause ob- servable evaporation signals in other systems, making them suitable for follow-up observations.
In the next generation of space observatories, X-ray transmission spectroscopy of an exoplanet’s atmosphere will be possible, allowing for a detailed look into the atmospheric composition of these planets. In chapter 7, we model sample spectra using a toy model of an exoplanetary atmosphere to predict what exoplanet transit observations with future X-ray missions such as Athena will look like. We then estimate the observable X-ray transmission spectrum for a typical Hot Jupiter-type exoplanet, giving us insights into the advances in X-ray observations of exoplanets in the decades to come.