You can simply rely on communities for a robust characterization of stances

نویسندگان

چکیده

We show that the structure of communities in social me- dia provides robust information for weakly supervised approaches to assign stances tweets. Using as seed SemEval 2016 Stance Detection Task annotated data, we retrieved a high number topically related then propagated from manually an- notated tweets and thus obtained bigger training corpus. Classifiers trained with this bigger, dataset reach similar or better performance than those seed. In addition, they are more respect common manual annotator errors biases have arguably coverage smaller datasets. Weakly approaches, most notably self- supervision, commonly suffer error propagation. Interestingly, seem provide constrains particular, classifiers incorporate community struc- ture class imbalance. Additionally, is straightforward, transparent ap- proach, using standard tools pipelines, cheaper faster methods like crowd sourcing notations. Thus it facilitates adoption, interpretability accountability.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Can we Rely on SQL?

It is important for any data language that it enables many people to derive correct information from a databases in a simple, effective way with predictable performance. In an analysis is shown that SQL cannot fulfill these essential preconditions. It is caused by insufficient structural semantics of the relational model and the related operations in SQL. This follows from a comparison with res...

متن کامل

a study on thermodynamic models for simulation of 1,3 butadiene purification columns

attempts have been made to study the thermodynamic behavior of 1,3 butadiene purification columns with the aim of retrofitting those columns to more energy efficient separation schemes. 1,3 butadiene is purified in two columns in series through being separated from methyl acetylene and 1,2 butadiene in the first and second column respectively. comparisons have been made among different therm...

Can we rely on retrospective pain assessments?

The objective of this paper was to study whether subjects in a workplace setting are able to assess the intensity of musculoskeletal pain retrospectively for a period of 3 months. The intensity of average pain and maximum pain in eight anatomic regions was assessed on a numeric rating scale. The results of 12 consecutive weekly pain recordings were compared with a final retrospective assessment...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Proceedings of the ... International Florida Artificial Intelligence Research Society Conference

سال: 2021

ISSN: ['2334-0762', '2334-0754']

DOI: https://doi.org/10.32473/flairs.v34i1.128515