Machine Learning (ML) is increasingly applied in real-life scenarios, raising concerns about bias in automatic decision making. We focus on bias as a notion of opinion exclusion, that stems from the direct application of traditional ML pipelines to infer subjective properties. We argue that such ML systems should be evaluated with subjectivity and bias in mind. Considering the lack of evaluation standards yet to create evaluation benchmarks, we propose an initial list of specifications to define prior to creating evaluation datasets, in order to later accurately evaluate the biases. With the example of a sentence toxicity inference system, we illustrate how the specifications support the analysis of biases related to subjectivity. We highlight difficulties in instantiating these specifications and list future work for the crowdsourcing community to help the creation of appropriate evaluation datasets.
Technische Universität Berlin（柏林工业大学）；DFKI Projektbüro Berlin（德国人工智能研究中心柏林项目办公室）
This paper presents TextComplexityDE, a dataset consisting of 1000 sentences in German language taken from 23 Wikipedia articles in 3 different article-genres to be used for developing text-complexity predictor models and automatic text simplification in German language. The dataset includes subjective assessment of different text-complexity aspects provided by German learners in level A and B. In addition, it contains manual simplification of 250 of those sentences provided by native speakers and subjective assessment of the simplified sentences by participants from the target group. The subj…
Language is ambiguous; many terms and expressions can convey the same idea. This is especially true in creative practice, where ideas and design intents are highly subjective. We present a dataset, Ambiguous Descriptions of Art Images (ADARI), of contemporary workpieces, which aims to provide a foundational resource for subjective image description and multimodal word disambiguation in the context of creative practice. The dataset contains a total of 240k images labeled with 260k descriptive sentences. It is additionally organized into sub-domains of architecture, art, design, fashion, furnitu…
Harbin Institute of Technology（哈尔滨工业大学）；Tencent CSIG（腾讯云与智慧产业事业群）
With the recent proliferation of the use of text classifications, researchers have found that there are certain unintended biases in text classification datasets. For example, texts containing some demographic identity-terms (e.g., "gay", "black") are more likely to be abusive in existing abusive language detection datasets. As a result, models trained with these datasets may consider sentences like "She makes me happy to be gay" as abusive simply because of the word "gay." In this paper, we formalize the unintended biases in text classification datasets as a kind of selection bias from the no…
Can Deception Detection Go Deeper? Dataset, Evaluation, and Benchmark for Deception Reasoning
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；Shanghai Jiao Tong University（上海交通大学）；Inner Mongolia University（内蒙古大学）；Tsinghua University（清华大学）；Peking University（北京大学）
In this paper, we introduce a new Czech subjectivity dataset of 10k manually annotated subjective and objective sentences from movie reviews and descriptions. Our prime motivation is to provide a reliable dataset that can be used with the existing English dataset as a benchmark to test the ability of pre-trained multilingual models to transfer knowledge between Czech and English and vice versa. Two annotators annotated the dataset reaching 0.83 of the Cohen's \k{appa} inter-annotator agreement. To the best of our knowledge, this is the first subjectivity dataset for the Czech language. We also…
MIT Computer Science and Artificial Intelligence Laboratory（麻省理工学院计算机科学与人工智能实验室）；American University of Beirut（贝鲁特美国大学）；Qatar University（卡塔尔大学）
Sentiment analysis is a highly subjective and challenging task. Its complexity further increases when applied to the Arabic language, mainly because of the large variety of dialects that are unstandardized and widely used in the Web, especially in social media. While many datasets have been released to train sentiment classifiers in Arabic, most of these datasets contain shallow annotation, only marking the sentiment of the text unit, as a word, a sentence or a document. In this paper, we present the Arabic Sentiment Twitter Dataset for the Levantine dialect (ArSenTD-LEV). Based on findings fr…
Due to flourish of the Web 2.0, web opinion sources are rapidly emerging containing precious information useful for both customers and manufactures. Recently, feature based opinion mining techniques are gaining momentum in which customer reviews are processed automatically for mining product features and user opinions expressed over them. However, customer reviews may contain both opinionated and factual sentences. Distillations of factual contents improve mining performance by preventing noisy and irrelevant extraction. In this paper, combination of both supervised machine learning and rule-b…
Text Detoxification as Style Transfer in English and Hindi
Charles University, Faculty of Mathematics and Physics（查理大学数学与物理学院）；Panlingua Language Processing LLP（Panlingua语言处理有限责任合伙公司）；Insight SFI Centre for Data Analytics, DSI, University of Galway（爱尔兰高威大学Insight SFI数据分析中心）
Harvard Medical School（哈佛医学院）；Elder Research, Inc.（埃尔德研究公司）；Clarkson University（克拉克森大学）
The rapid growth of text data has motivated the development of machine-learning based automatic text summarization strategies that concisely capture the essential ideas in a larger text. This study aimed to devise an extractive summarization method for A-133 Single Audits, which assess if recipients of federal grants are compliant with program requirements for use of federal funding. Currently, these voluminous audits must be manually analyzed by officials for oversight, risk management, and prioritization purposes. Automated summarization has the potential to streamline these processes. Analy…
Sofia University “St Kliment Ohridski”（索非亚大学“圣克莱门特·奥赫里德斯基”）；National University of Singapore（新加坡国立大学）
We describe a novel language-independent approach to the task of determining the polarity, positive or negative, of the author's opinion on a specific topic in natural language text. In particular, weights are assigned to attributes, individual words or word bi-grams, based on their position and on their likelihood of being subjective. The subjectivity of each attribute is estimated in a two-step process, where first the probability of being subjective is calculated for each sentence containing the attribute, and then these probabilities are used to alter the attribute's weights for polarity c…
