27 research outputs found
Development and evaluation of data correction methods for the PETRRA positron camera
No full textEThOS - Electronic Theses Online ServiceGBUnited Kingdo
Software for "Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior"
No full textThis repository consists of the custom external platform for the annotation process of CrowdFlower, used on the "Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior" paper, published in ICWSM 2018. Full text of the paper can be found here:
Please cite the paper in any published work that uses any of these resources.
@article{founta2018large,
title={Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior},
author={Founta, Antigoni-Maria and Djouvas, Constantinos and Chatzakou, Despoina and Leontiadis, Ilias and Blackburn, Jeremy and Stringhini, Gianluca and Vakali, Athena and Sirivianos, Michael and Kourtellis, Nicolas},
journal={arXiv preprint arXiv:1802.00393},
year={2018}
}
For any further questions contact a.m.founta at gmail dot com.</p
Rensch’s rule: linking intraspecific to evolutionary allometry
No full textSexual dimorphism describes phenotypic differences between the sexes; the most prominent of which is sexual size dimorphism (SSD). Rensch's Rule (RR) is an allometric trend in which SSD increases in male-larger taxa and decreases in female-larger ones. Covariation between a trait and overall size within and across species can both be affected by sexual and natural selection. Thus, intraspecific allometric variation could influence the expression of RR. Here we used computer simulations to dissect how RR emerges under specific allometric patterns of intraspecific sexual differentiation in a trait. We found that sexual differentiation in static allometric slopes is the main determinant of RR. Based on our findings, RR and its converse can manifest in both body size and other traits. As a realistic showcase, we also examined RR and static allometry of different body parts in Mediterranean green lizards to establish whether intraspecific and evolutionary allometry are linked. Here, we identified RR and its converse for different traits, where the amount of sexual differentiation in static allometric slopes within species had a significant contribution to RR. Integrating the simulations and the empirical case we corroborate that sexual differentiation in static allometric slopes is a major parameter affecting evolutionary allometry.This is a manuscript of an article published as Carolina Reyes-Puig, Dean C Adams, Urtzi Enriquez-Urzelai, Antigoni Kaliontzopoulou, Rensch’s rule: linking intraspecific to evolutionary allometry, Evolution, 2023;, qpad172, https://doi.org/10.1093/evolut/qpad172. © The Author(s) 2023. Posted with permission
Dataset for "Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior"
No full textDataset for the publication "Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior". Antigoni-Maria Founta, Constantinos Djouvas, Despoina Chatzakou, Ilias Leontiadis, Jeremy Blackburn, Gianluca Stringhini, Athena Vakali, Michael Sirivianos and Nicolas Kourtellis. International AAAI Conference on Web and Social Media (ICWSM), 2018.
The dataset provided here includes an updated version of the original dataset, with ~100k tweets annotated using the CrowdFlower platform:
hatespeech_labels.csv: contains ~100k rows, where every row consists of a unique Tweet ID and its associated majority annotation
UPDATE: It has come to our understanding that a number of the tweets are not available anymore for download on Twitter. Therefore, upon request, we can provide one more file with the full ~100k tweet text and their associated majority labels. The tweets are shuffled so that there is no connection between tweet IDs and texts (in order to be aligned with the T&C of Twitter).
To obtain the file contact a.m.founta at gmail dot com AND antonis26papa at gmail dot com.
Please cite the paper in any published work that uses any of these resources.
@inproceedings{founta2018large,
title={Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior},
author={Founta, Antigoni-Maria and Djouvas, Constantinos and Chatzakou, Despoina and Leontiadis, Ilias and Blackburn, Jeremy and Stringhini, Gianluca and Vakali, Athena and Sirivianos, Michael and Kourtellis, Nicolas},
booktitle={11th International Conference on Web and Social Media, ICWSM 2018},
year={2018},
organization={AAAI Press}
}
For any further questions contact a.m.founta at gmail dot com.
Publication DOI: https://doi.org/10.5281/zenodo.1443348
Github: https://github.com/ENCASEH2020/hatespeech-twitter
The updated version of this Dataset is here: https://zenodo.org/record/2657374#.XMrDIY4zaUk</p
Bricks as Spatial Sound Modulators: Towards tuning the geometry
No full textThe current research is part of a trans-disciplinary discussion on the long-lasting impact of the acoustic environment on cognitive performance, literacy and auditory perception for preschool children, aiming to underlie the increasing necessity to perfectly tune every-day facilities such as schools and work spaces, rather than only performance auditoriums. The main focus is placed on improving speech intelligibility, which is the task of recognizing the message from the din, by evaluating the speech-in-noise perception. The research is based on a promising hypothesis that hybrid devices should allow for low-frequency absorption and high-frequency diffusion which correspond to certain properties of vowels and consonants respectively. Given the transition to the modern aesthetics of the flat wall, compared to the inherent complexity of surface ornamentation in the past, there is a growing demand to provide acoustic solutions that allow for a higher level of complexity towards an improved acoustic environment. The effect of periodicity on a uniform scattered distribution is investigated by examining different geometrical patterns, aiming to convert traditional materials into sound controlling devices. Customized and mass-produced versions are developed,arguing whether a lower level of complexity can meet the same goals if certain low-level rules with high level of self-organization are applied, introducing the idea of simplicity and architectural smartness by preserving the economy of means for the end users. Last but not least, since speech transmission is related to spatial context and surface properties, both the macro-scale and micro-scale are examined. The study envisions an open-source toolbox for designers for generating design alternatives providing a deeper understanding on how periodicity affects the acoustics of a given space. Since periodic surfaces reduce the complexity of the reflected sound field, designers are encouraged to focus on a-periodic patterns aiming for a non-repetitive sequence via the geometry articulation.Further it suggests the manipulation of certain parameters, such as period,depth-to-width ratio and well width, towards an integrated workflow to achieve an optimized acoustic design. These alternatives can then be used as input fora software similar to Pachyderm Acoustics plug-in for Grasshopper, in order to get a glimpse which of those showcase a high scattering performance within the relevant frequency range. This should for no reason substitute the necessity for a thorough acoustical study as a next step but would however, involve designers more actively in the initial process of form-finding, while gaining awareness of the impact of their design choices on the acoustic environment. Architecture, Urbanism and Building Sciences | Building Technology | Sustainable Desig
Adaptive fabric façade for a high-rise in Paris
No full textThe topic of this graduation project is the design of an adaptive fabric façade for a high-rise in Paris. The stimulus for this choice was the turn of nowadays towards lightweight materials and constructions, a trend that is growing tremendously. Glass façades always remain popular and are frequently used in all kinds of buildings. For instance, a great quantity of glass is used in curtain walls. Therefore, following the need of research for lightweight materials, alternatives for glass are being investigated and membranes can be an interesting solution. They offer a number of possibilities and have many advantages, such as a variety of transparency levels and the fact that they are extremely lightweight materials. Furthermore, this graduation project concerns an office building, which is a rather demanding building type, as it requires a building envelope that offers a high level of indoor comfort, in terms of thermal and acoustical insulation, as well as efficient solar control. Thus, taking all the above into account and the global growing interest towards membranes, which are mostly used as roof components until now, the decision of exploring the potential of such materials as the main part of a building envelope for an open-plan office building in Paris was taken. The graduation research has an experimental character, as it tackles several issues of a possible membrane façade element and it tries to answer the questions that are being raised throughout of the development of this thesis. Firstly, a literature research is conducted on the available materials, their physical and mechanical properties, their production and connection methods, their advantages and drawbacks in general. Subsequently, design aspects are taken into consideration concerning the building envelope and physical models are constructed in order to visualise the design. Furthermore, hand calculations are conducted for the thermal and acoustical performance of the under study membrane façade element and then software simulations are made in order to validate the results. Moreover, an actual acoustical test is attempted with several difficulties, though, that will be described later on. Finally, conclusions are drawn concerning the whole research and design. As it is evident in the end, although the proposed solution is proved to be rather successful, there is still space for improvement, as well as further research and development.Sustainable graduation studioArchitecture, Urbanism and Building Science
Invalid Dataset for "Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior"
No full textThis dataset is invalid. The updated version of this Dataset is here: https://zenodo.org/record/3678559#.Xl9-Ji97FhE
Dataset for the "Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior" paper, published in ICWSM 2018. The full text of the paper can be found here.
The dataset provided here includes an updated version of the original dataset, with ~100k tweets annotated using the CrowdFlower platform:
hatespeech_labels.csv: contains ~100K rows, where every row consists of a unique Tweet ID and its according to majority annotation
UPDATE: It has come to our understanding that a number of the tweets are not available anymore for download on Twitter. Therefore, under request, we can provide one more file with the full ~100K tweet text, their associated majority label, and the number of votes for the majority label. The tweets are shuffled so that there is no connection between tweet IDs and texts (in order to be in line with the T&C of Twitter). To obtain the file contact the authors through email.
Please cite the paper in any published work that uses any of these resources.
@inproceedings{founta2018large,
title={Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior},
author={Founta, Antigoni-Maria and Djouvas, Constantinos and Chatzakou, Despoina and Leontiadis, Ilias and Blackburn, Jeremy and Stringhini, Gianluca and Vakali, Athena and Sirivianos, Michael and Kourtellis, Nicolas},
booktitle={11th International Conference on Web and Social Media, ICWSM 2018},
year={2018},
organization={AAAI Press}
}
For any further questions contact a.m.founta at gmail dot com AND markos.charalambous at eecei.cut.ac.c
Restricted Dataset for "Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior"
No full textRestricted Dataset for the "Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior" paper, published in ICWSM 2018. The full text of the paper can be found here. The Public version of the dataset can be found here
hatespeech_text_label_vote_RESTRICTED_100K.csv: contains ~100K raws with tweet text, the associated majority label, and the number of votes for the majority label. The tweets are shuffled so that there is no connection between tweet IDs and texts (in order to be in line with the T&C of Twitter).
retweets.csv: contains ~2K rows, where every row consists of the row number in the hatespeech_text_label_vote_RESTRICTED_100K.csv file which is the first occurrence of a Tweet text followed by comma-separated row numbers of all other occurrences of the same Tweet text in the same file. There are ~8K other occurrences due to retweets.
Please cite the paper in any published work that uses any of these resources.
@inproceedings{founta2018large,
title={Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior},
author={Founta, Antigoni-Maria and Djouvas, Constantinos and Chatzakou, Despoina and Leontiadis, Ilias and Blackburn, Jeremy and Stringhini, Gianluca and Vakali, Athena and Sirivianos, Michael and Kourtellis, Nicolas},
booktitle={11th International Conference on Web and Social Media, ICWSM 2018},
year={2018},
organization={AAAI Press}
}
For any further questions contact a.m.founta at gmail dot com AND markos.charalambous at eecei dot cut dot ac dot c
