Proceedings of the LREC 2020 Workshop on "Citizen Linguistics in Language Resource Development"
This paper introduces the citizen science platform, LanguageARC, developed within the NIEUW (Novel Incentives and Workflows) project supported by the National Science Foundation under Grant No. 1730377. LanguageARC is a community-oriented online platform bringing together researchers and “citizen linguists” with the shared goal of contributing to linguistic research and language technology development. Like other Citizen Science platforms and projects, LanguageARC harnesses the power and efforts of volunteers who are motivated by the incentives of contributing to science, learning and discovery, and belonging to a community dedicated to social improvement. Citizen linguists contribute language data and judgments by participating in research tasks such as classifying regional accents from audio clips, recording audio of picture descriptions and answering personality questionnaires to create baseline data for NLP research into autism and neurodegenerative conditions. Researchers can create projects on Language ARC without any coding or HTML required using our Project Builder Toolkit.
Language resources are a major ingredient for the advancement of language technologies. Citizen linguistics can help to create language resources and annotate language resources, not only for the improvement of language technologies, such as machine translation but also for the advancement of linguistic research. The (language) resources covered in this article are a corpus related to the Question of the Month project strand, which was initially aimed at co-creation in citizen linguistics and a partially annotated database of pictures of written text in different languages found in the public sphere. The number of participants in these project strands differed significantly. Especially those activities that were related to data collection (and analysis) had a significantly higher number of contributions per participant. This especially held true for the activities with (prize) incentives. Nevertheless, the activities of the Question of the Month could reach a higher number of participants, even after the co-creation approach was no longer followed. In addition, the Question of the Month brought research gaps and new knowledge to light and challenged existing paradigms and practices. These are especially important for the advancement of scholarly research. Citizen linguistics can help gather and analyze linguistic data, including language resources, in a short period of time. Thus, it may help increase the access to and availability of language resources.
Labelling, or annotation, is the process by which we assign labels to an item with regards to a task. In some Artificial Intelligence problems, such as Computer Vision tasks, the goal is to obtain objective labels. However, in problems such as text and sentiment analysis, subjective labelling is often required. More so when the sentiment analysis deals with actual emotions instead of polarity (positive/negative) . Scientists employ human experts to create these labels, but it is costly and time consuming. Crowdsourcing enables researchers to utilise non-expert knowledge for scientific tasks. From image analysis to semantic annotation, interested researchers can gather a large sample of answers via crowdsourcing platforms in a timely manner. However, non-expert contributions often need to be thoroughly assessed, particularly so when a task is subjective. Researchers have traditionally used ‘Gold Standard’, ‘Thresholding’ and ‘Majority Voting’ as methods to filter non-expert contributions. We argue that these methods are unsuitable for subjective tasks, such as lexicon acquisition and sentiment analysis. We discuss subjectivity in human centered tasks and present a filtering method that defines quality contributors, based on a set of objectively infused terms in a lexicon acquisition task. We evaluate our method against an established lexicon, the diversity of emotions - i.e. subjectivity- and the exclusion of contributions. Our proposed objective evaluation method can be used to assess contributors in subjective tasks that will provide domain agnostic, quality results, with at least 7% improvement over traditional methods.
Crowdsourcing approaches provide a difficult design challenge for developers. There is a trade-off between the efficiency of the task to be done and the reward given to the user for participating, whether it be altruism, social enhancement, entertainment or money. This paper explores how crowdsourcing and citizen science systems collect data and complete tasks, illustrated by a case study from the online language game-with-a-purpose Phrase Detectives. The game was originally developed to be a constrained interface to prevent player collusion, but subsequently benefited from posthoc analysis of over 76k unconstrained inputs from users. Understanding the interface design and task deconstruction are critical for enabling users to participate in such systems and the paper concludes with a discussion of the idea that social networks can be viewed as form of citizen science platform with both constrained and unconstrained inputs making for a highly complex dataset.
Abstract Meaning Representations (AMRs), a syntax-free representation of phrase semantics are useful for capturing the meaning of a phrase and reflecting the relationship between concepts that are referred to. However, annotating AMRs are time consuming and expensive. The existing annotation process requires expertly trained workers who have knowledge of an extensive set of guidelines for parsing phrases. In this paper, we propose a cost-saving two-step process for the creation of a corpus of AMR-phrase pairs for spatial referring expressions. The first step uses non-specialists to perform simple annotations that can be leveraged in the second step to accelerate the annotation performed by the experts. We hypothesize that our process will decrease the cost per annotation and improve consistency across annotators. Few corpora of spatial referring expressions exist and the resulting language resource will be valuable for referring expression comprehension and generation modeling.
We report on a web-based resource for conducting intercomprehension experiments with native speakers of Slavic languages and present our methods for measuring linguistic distances and asymmetries in receptive multilingualism. Through a website which serves as a platform for online testing, a large number of participants with different linguistic backgrounds can be targeted. A statistical language model is used to measure information density and to gauge how language users master various degrees of (un)intelligibilty. The key idea is that intercomprehension should be better when the model adapted for understanding the unknown language exhibits relatively low average distance and surprisal. All obtained intelligibility scores together with distance and asymmetry measures for the different language pairs and processing directions are made available as an integrated online resource in the form of a Slavic intercomprehension matrix (SlavMatrix).
This study uses crowdsourcing through LanguageARC to collect data on levels of accuracy in the identification of speakers’ ethnicities. Ten participants (5 US; 5 South-East England) classified lexically identical speech stimuli from a corpus of 227 speakers aged 18-33yrs from South-East England into the main “ethnic” groups in Britain: White British, Black British and Asian British. Firstly, the data reveals that there is no significant geographic proximity effect on performance between US and British participants. Secondly, results contribute to recent work suggesting that despite the varying heritages of young, ethnic minority speakers in London, they speak an innovative and emerging variety: Multicultural London English (MLE) (e.g. Cheshire et al., 2011). Countering this, participants found perceptual linguistic differences between speakers of all 3 ethnicities (80.7% accuracy). The highest rate of accuracy (96%) was when identifying the ethnicity of Black British speakers from London whose speech seems to form a distinct, perceptual category. Participants also perform substantially better than chance at identifying Black British and Asian British speakers who are not from London (80% and 60% respectively). This suggests that MLE is not a single, homogeneous variety but instead, there are perceptual linguistic differences by ethnicity which transcend the borders of London.
LanguageARC is a portal that offers citizen linguists opportunities to contribute to language related research. It also provides researchers with infrastructure for easily creating data collection and annotation tasks on the portal and potentially connecting with contributors. This document describes LanguageARC’s main features and operation for researchers interested in creating new projects and or using the resulting data.