Predictive Linguistics

---

Predictive Linguistics (PL)

Predictive Linguistics (PL) is an emerging discipline that studies the interrelation between linguistic markers and action, considered as an intentional, purposive, conscious, and subjectively meaningful activity. A primary concern of PL is to analyze the way language may predict human actions and explain the relationship between time and action within language, notably how a future action is expressed in present language. The discipline is mainly concerned with the mechanisms by which future actions and mind states are processed and represented in natural language. Research makes use of linguistics, neuroscience, cognitive science, and information science to analyze current and historical data to make predictions about future actions or otherwise unknown events.

Foundation

Predictive linguistics was founded by Prof. Mathieu Guidere, PhD in linguistics from The Sorbonne (University of Paris IV, France). Guidere coined the term "predictive linguistics" in 2006 while he was a professor at the French Military Academy of Saint-Cyr and director of the Strategic Information Analysis Unit.. In 2015, he published a book in French that summarizes his work on predictive linguistics: "La Linguistique prédictive: de la cognition à l’action", one of his numerous writings on language and security. Guidere had already previously written many publications discussing the role of linguistic markers in the prediction of violent actions. Some of these previous publications include “Rich Language Analysis for Counterterrorism” and the detection of suicide bombing plans through the predictive analysis of suicide bombers’ notes and “testaments”. In 2011, he published the paper "Computational Methods for Clinical Applications: An Introduction", in which he also coined the term "cognitive computing”. Soon after, he developed, with Prof. Newton Howard, LXIO: the Mood Detection RoboPsych. By the end of 2012, the field had attracted the attention of many people and started to grow.

Prediction in language

Prediction in language is different from linguistic prediction. It aims at forecasting what would happen in the future based on a rigorous analysis of linguistic data and markers in the present. Linguistic predictability is established both qualitatively and quantitatively: experimental methods indicate that combining statistical word processing and automatic discourse analysis within the context formed by words may, under certain conditions, enable the prediction of upcoming actions and mind states. Further, predictability has been shown to be grounded on semantic and pragmatic markers. Limitations on predictability could be caused by a lack of information on time or action within linguistic data. The combination of predictive linguistics and predictive modeling enables researchers, analysts, physicians, and decision-makers to aggregate and analyze disparate types of data, recognize patterns and trends within that data, and make more informed decisions in an effort to preemptively alter future outcomes.

Predictive linguistics and Big Data

Predictive linguistics is used to analyze the Big data that are so large and complex that they become awkward to work with using traditional database management tools. Examples of linguistic big data sources include web logs and social networks. Predictive linguistics is the core of text analytics for processing big data. It also enables running predictive algorithms on streaming audio or text data. Today, exploring big data and using predictive linguistics is within reach of more organizations than ever before, and new methods capable of handling such datasets are being proposed. In order to analyze Big data, researchers in predictive linguistics use Machine learning, a branch of artificial intelligence, that enables computers to learn. Today, since it includes a number of quantitative and qualitative methods for data analysis, it finds application in a wide variety of fields, especially medical diagnostics. In these applications, machine learning techniques emulate human cognition and learn from training examples to predict future actions and events.

Current uses of predictive linguistics

Predictive linguistics is used in healthcare, security, telecommunications, actuarial science, marketing, insurance, financial services, and other fields. Below, we outline two areas where it has shown positive impact in recent years.

Healthcare Predictive linguistics is used to analyze the vast amounts of clinical data that are now computable. Researchers correlate this data with other datasets (e.g. electronic health records) in order to conduct predictive analytics. For instance, psychotherapists and mental health professionals use predictive linguistics to transform audio recorded and transcribed data into medical knowledge in order to make better decisions faster in the area of mental disorders identification and therapy optimization. Experts use predictive linguistics in healthcare primarily to determine which patients are at risk of developing certain mental disorders, like schizophrenia, paranoia, mood disorders, and personality disorders. Additionally, sophisticated clinical decision support systems incorporate predictive linguistics to support medical diagnosis and disease progression forecasting in many neurodegenerative disorders like Alzheimer’s.

Other uses Predictive linguistics is also used in security systems to integrate complex data about the future actions and plans of individuals based on the automatic analysis of their verbal interactions and audio recordings. It exploits patterns found in linguistic data to identify actions and mind states. Predictive linguistics captures relationships among many semantic data to allow assessment of risk or potential threat associated with a particular set of actions expressed in discourse.

Tools

Predictive linguistics tools’ development requires advanced skills in natural language processing, but it is not restricted to computational linguists. As the field is emerging, there are very few tools available in the marketplace that help with the execution of predictive linguistics, and they are designed for the expert practitioner.

Criticism

Some experts are skeptical about algorithms' abilities to predict the future based on what people say or write. Predictive linguistics might not tell exactly what someone will do next, but it can analyze their intention. There are definitely a lot of unpredictable variables, but linguistic markers remain the most predictable because they express conceptions, perceptions, and actions.

See also

• Natural language processing (NLP)
• Data mining
• Predictive modelling
• Predictive analytics
• Predictive inference
• Machine learning

Further reading

• Guidere M. (2015), La Linguistique prédictive : de la cognition à l’action, Paris: Editions L’Harmattan.
• Al-Qaeda's Noms de Guerre, in Defense Concepts, Vol.1, Edition 3, Fall 2006, pp. 6-16.
• Guidere M. et al. (2009), Rich Language Analysis for Counterterrorism, in Computational Methods for Counterterrorism pp 109-120.
• Guidere M. & Howard N. (2011), Computational Methods for Clinical Applications: An Introduction. Functional Neurology, Rehabilitation, and Ergonomics. 2011, 1:237-250.
• Guidere M. & Howard N., (2012). LXIO The Mood Detection Robopsych. The Brain Sciences Journal, 1(1), 98-109.
• Guidere M. (2019), Predictive Linguistics Applied to Suicide Notes Posted Online by Suicide Attempters and Suicide Completers, 9th World Congress on Mental Health, Psychiatry and Well-being, New York, 20-21 March, 2019.
• Guidere M. & Howard N., (2012). LXIO The Mood Detection Robopsych. The Brain Sciences Journal, 1(1), 98-109.

External links

• https://www.tolerance.ca/Article.aspx?ID=71064&L=fr
• https://www.researchgate.net/publication/249012184_Guidere_How_to_decode_Defense_Concepts
• https://link.springer.com/chapter/10.1007/978-3-642-01141-2_7
• https://www.amazon.fr/Martyrs-dAl-Qaida-Mathieu-Guid%C3%A8re/dp/2842743695
• http://braindomain.org/wp-content/uploads/Howard-Guidere-2011-LXio-The-Mood-Detection-Robopsych.pdf Archived 2016-04-24 at the Wayback Machine
• https://annualmentalhealth.psychiatryconferences.com/events-list/mental-health-and-wellbeing
• http://braindomain.org/wp-content/uploads/Howard-Guidere-2011-LXio-The-Mood-Detection-Robopsych.pdf Archived 2016-04-24 at the Wayback Machine

Predictive Linguistics

This article "Predictive Linguistics" is from Wikipedia. The list of its authors can be seen in its historical and/or the page Edithistory:Predictive Linguistics. Articles copied from Draft Namespace on Wikipedia could be seen on the Draft Namespace of Wikipedia and not main one.

---