Prediction of Creole Emergence in Spatial Language Dynamics

Nakamura, Makoto; Hashimoto, Takashi; Tojo, Satoshi

doi:10.1007/978-3-642-00982-2_52

Cited by 3 publications

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Research on creoles is more common in the field of language evolution than in NLP. In particular, work on creoles in this field typically focuses on their computational modeling, their emergence (Nakamura et al, 2009), and their evolution (Jansson et al, 2015;Furman and Nitschke, 2020). Other creole modeling efforts in this space may be more tailored towards specific linguistic insights (Parkvall, 2008).…”

Section: Computational Research On Language Evolutionmentioning

confidence: 99%

On Language Models for Creoles

Lent¹,

Bugliarello²,

Lhoneux³

et al. 2021

Proceedings of the 25th Conference on Computational Natural Language Learning

View full text Add to dashboard Cite

Creole languages such as Nigerian Pidgin English and Haitian Creole are under-resourced and largely ignored in the NLP literature. Creoles typically result from the fusion of a foreign language with multiple local languages, and what grammatical and lexical features are transferred to the creole is a complex process (Sessarego, 2020). While creoles are generally stable, the prominence of some features may be much stronger with certain demographics or in some linguistic situations (Winford, 1999;Patrick, 1999). This paper makes several contributions: We collect existing corpora and release models for Haitian Creole, Nigerian Pidgin English, and Singaporean Colloquial English. We evaluate these models on intrinsic and extrinsic tasks. Motivated by the above literature, we compare standard language models with distributionally robust ones and find that, somewhat surprisingly, the standard language models are superior to the distributionally robust ones. We investigate whether this is an effect of overparameterization or relative distributional stability, and find that the difference persists in the absence of over-parameterization, and that drift is limited, confirming the relative stability of creole languages.

show abstract

Section: Computational Research On Language Evolutionmentioning

confidence: 99%

On Language Models for Creoles

Lent¹,

Bugliarello²,

Lhoneux³

et al. 2021

Proceedings of the 25th Conference on Computational Natural Language Learning

View full text Add to dashboard Cite

show abstract

Modeling the Emergence of a New Language: Naming Game with Hybridization

Pucci¹,

Gravino

Servedio

2014

Self-Organizing Systems

View full text Add to dashboard Cite

In recent times, the research field of language dynamics has focused on the investigation of language evolution, dividing the work in three evolutive steps, according to the level of complexity: lexicon, categories and grammar. The Naming Game is a simple model capable of accounting for the emergence of a lexicon, intended as the set of words through which objects are named. We introduce a stochastic modification of the Naming Game model with the aim of characterizing the emergence of a new language as the result of the interaction of agents. We fix the initial phase by splitting the population in two sets speaking either language A or B. Whenever the result of the interaction of two individuals results in an agent able to speak both A and B, we introduce a finite probability that this state turns into a new idiom C, so to mimic a sort of hybridization process. We study the system in the space of parameters defining the interaction, and show that the proposed model displays a rich variety of behaviours, despite the simple mean field topology of interactions. Emergence of a lexicon as a languageThe modeling activity of language dynamics aims at describing language evolution as the global effect of the local interactions between individuals in a population of N agents, who tend to align their verbal behavior locally, by a negotiation process through which a successful communication is achieved [1,2]. In this framework, the emergence of a particular communication system is not due to an external coordination, or a common psychological background, but it simply occurs as a convergence effect in the dynamical processes that start from an initial condition with no existing words (agents having to invent them), or with no agreement.Our work is based on the Naming Game (NG) model, and on its assumptions [3]. In Fig. 1 we recall the NG basic pairwise interaction scheme. A fundamental assumption of NG is that vocabulary evolution associated to every * The final publication will be available at http://www.springer.com/lncs SUCCESSFigure 1: Naming Game interaction scheme. A speaker and a hearer are picked up randomly. The speaker utters a word chosen randomly in his vocabulary. Failure: the hearer does not know the uttered word and he simply adds it to his vocabulary. Success: the hearer knows the uttered word and both agents by agreement delete all the words in their vocabularies except the uttered word one. single object is considered independent. This lets us simplify the evolution of the whole lexicon as the evolution of the set of words associated to a single object, equally perceived in the sensorial sphere by all agents.The simplicity of NG in describing the emergence of a lexicon also relies on the fact that competing words in an individual vocabulary are not weighted, so that they can be easily stored or deleted [4]. In fact it turns out that for convergence to a single word consensus state, the weights are not necessary as it was supposed by the seminal work in this research field [3]. Every agent is a bo...

show abstract