Tekstgenres analyseren op lexicale complexiteit met T‑Scan

Maat, Henk Pander; Dekker, Natashe Lemos

doi:10.5117/tvt2016.3.pand

Cited by 6 publications

(8 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Perhaps focusing on the presence of nouns, by counting articles and prepositions (Tausczik and Pennebaker, 2010) is not a very valid way of measuring the concreteness of text. Nouns per se are not concrete; they can have different degrees of concreteness (Pander Maat and Dekker, 2016). For example, words like 'stove', 'pan', and 'meat' are considered to be concrete words, whereas words such as 'additives' and 'cereal products' are seen as more abstract -yet, all are nouns.…”

Section: Discussionmentioning

confidence: 99%

Promoting trust through linguistic features of provider profiles in the sharing economy

Huurne¹,

Moons²,

Ronteltap³

et al. 2021

IJEB

View full text Add to dashboard Cite

Trust between providers and consumers in the sharing economy are crucial to complete transactions successfully. From a consumer's perspective, a provider's profile is an important source of information for judging trustworthiness, because it contains multiple trust cues. However, the effect of a provider's self-description on perceived trustworthiness is still poorly understood. We examine how the linguistic features of a provider's self-description predict perceived trustworthiness. To determine the perceived trustworthiness of 259 profiles, real consumers on a Dutch sharing platform rated these profiles for trustworthiness. The results show that profiles were perceived as more trustworthy if they contained more words, more words related to cooking, and more words related to positive emotions. Also, a profile's perceived trustworthiness score correlated positively with the provider's actual sales performance. These findings indicate that a provider's self-description is a relevant signal to consumers, even though it seems easy to fake.

show abstract

Section: Discussionmentioning

confidence: 99%

Promoting trust through linguistic features of provider profiles in the sharing economy

Huurne¹,

Moons²,

Ronteltap³

et al. 2021

IJEB

View full text Add to dashboard Cite

show abstract

“…For example, word prevalence, entropy, perplexity, word count, particular character count, word probability, morpheme count, word frequencies, named entity recognition, sentence analysing features and many more. Some tooling uses more than 200 features [12], or even more than 400 formulas [5].…”

Section: Readability Toolingmentioning

confidence: 99%

“…T-Scan is a tool which provides 457 different kinds of measurement features for a Dutch document [17]. The 457 features are divided into nine groups.…”

Section: T-scanmentioning

confidence: 99%

See 1 more Smart Citation

Readability Metrics for Machine Translation in Dutch: Google vs. Azure & IBM

et al. 2023

View full text Add to dashboard Cite

This paper introduces a novel method to predict when a Google translation is better than other machine translations (MT) in Dutch. Instead of considering fidelity, this approach considers fluency and readability indicators for when Google ranked best. This research explores an alternative approach in the field of quality estimation. The paper contributes by publishing a dataset with sentences from English to Dutch, with human-made classifications on a best-worst scale. Logistic regression shows a correlation between T-Scan output, such as readability measurements like lemma frequencies, and when Google translation was better than Azure and IBM. The last part of the results section shows the prediction possibilities. First by logistic regression and second by a generated automated machine learning model. Respectively, they have an accuracy of 0.59 and 0.61.

show abstract

“…A straightforward method found the question. The method tokenised the sentences and selected the first sentence with a question mark as the leading question [22]. In the fifth step, the question, answer and documentation were combined in a JSON file, as an invalidated temporary dataset.…”

Section: Datasetmentioning

confidence: 99%

Exploring the Utility of Dutch Question Answering Datasets for Human Resource Contact Centres

et al. 2022

View full text Add to dashboard Cite

We explore the use case of question answering (QA) by a contact centre for 130,000 Dutch government employees in the domain of questions about human resources (HR). HR questions can be answered using personnel files or general documentation, with the latter being the focus of the current research. We created a Dutch HR QA dataset with over 300 questions in the format of the Squad 2.0 dataset, which distinguishes between answerable and unanswerable questions. We applied various BERT-based models, either directly or after finetuning on the new dataset. The F1-scores reached 0.47 for unanswerable questions and 1.0 for answerable questions depending on the topic; however, large variations in scores were observed. We conclude more data are needed to further improve the performance of this task.

show abstract

Tekstgenres analyseren op lexicale complexiteit met T‑Scan

Cited by 6 publications

References 23 publications

Promoting trust through linguistic features of provider profiles in the sharing economy

Promoting trust through linguistic features of provider profiles in the sharing economy

Readability Metrics for Machine Translation in Dutch: Google vs. Azure & IBM

Exploring the Utility of Dutch Question Answering Datasets for Human Resource Contact Centres

Contact Info

Product

Resources

About