Uncovering the differences in linguistic network dynamics of book and social media texts

Abstract

Complex network studies span a large variety of applications including linguistic networks. To investigate the differences in book and social media texts in terms of linguistic typology, we constructed both sequential and sentence collocation networks of book, Facebook and Twitter texts with undirected and weighted edges. The comparisons are performed using the basic parameters like average degree, modularity, average clustering coefficient, average path length, diameter, average link weight etc. We also presented the distribution graphs for node degrees, edge weights and maximum degree differences of the pairing nodes. The degree difference occurrences are furtherly detailed with the grayscale percentile plots with respect to the edge weights. We linked the network analysis with linguistic aspects like word and sentence length distributions. We concluded that linguistic typology demonstrates a formal usage in book that slightly deviates to informal in Twitter. Facebook interpolates between these media by the means of network parameters, while the informality of Twitter is mostly influenced by the character limitations.

A Comparative Study on Word Embeddings and Social NLP Tasks

“…These informal sources contain linguistic diversity, racial slurs and forms of profanity that do not exist in formal text (Türker et al, 2016). However, these social-media-based word embeddings have not been investigated for social NLP related tasks like cyberbullying detection and social bias analysis.…”

Proceedings of the Tenth International Workshop on Natural Language Processing for Social Media

“…We use the term "socialmedia-based" to describe those word embeddings. These informal sources contain linguistic diversity, racial slurs and forms of profanity that do not exist in formal text (Türker et al, 2016). However, these social-media-based word embeddings have not been investigated for social NLP related tasks like cyberbullying detection and social bias analysis.…”

VarioGram – Zaman serileri için renkli bir zaman-graf temsili

“…By the way, it includes more frequency components in the lower and less frequency components in the higher frequency range (Peng et al, 2021). An alternative framework for representing time series in a more structured format is powered by network science, which proposes that each complex interconnected system can be represented as a network (Baydilli et al, 2017;Demir & Türker, 2021;Türker et al, 2016;Türker & Sulak, 2018). Time series are converted into graph representations, which evaluate quantized amplitude levels as nodes and neighborhood between consecutive levels as edges between them (Lacasa et al, 2015).…”

“…Zaman serilerini daha yapılandırılmış bir formatta temsil etmek için alternatif bir çerçeve de, karmaşık olarak birbirine bağlı nesnelerden oluşan sistemlerin bir ağ formatında temsil edilebilmesidir (Baydilli et al, 2017;Demir & Türker, 2021;Türker et al, 2016;Türker & Sulak, 2018). Bu yöntem kullanılarak zaman serileri, nicelenmiş genlik seviyelerinin düğümler ve ardışık seviyeler arasındaki komşulukların ise kenarlar olarak değerlendirildiği graflar şeklinde temsil edilebilir (Lacasa et al, 2015).…”