From emotions to words and back: How AI can help you understand yourself better
(maind Monday 4/5)
Journaling is a popular practice for maintaining mental well-being, but it can be challenging to reflect on one’s emotions and experiences consistently. Natural language processing (NLP), a subfield of artificial intelligence (AI), offers new ways to enhance the journaling experience. You can read more about NLP in Elaine’s blog post “maind your mental health — where does the “ai” in “maind” come from?”.
“Natural language processing (NLP) is a field of artificial intelligence (AI) and machine learning (ML) that is concerned with using computers to process and analyse human language. Machine learning is generally divided into three areas: supervised learning, unsupervised learning, and reinforcement learning.”
– Elaine, Kasvu Labs’ Lead NLP Specialist
So, how could we use AI and NLP to enhance the journaling experience?
Keyword extraction and emotion analysis are powerful NLP tools that can be used to process journal entries. By extracting keywords and phrases from a text and analysing the emotions behind them, NLP can offer insights into the writer’s emotional state. Imagine receiving a journaling app reminder stating that words such as “jogging” and “forest” elicited feelings of “peace” and “happiness” — this can provide helpful feedback to encourage positive behaviours and activities.
Incorporating emojis can also be an effective way to enhance journaling. Emojis are fun and engaging emotion indicators that can help to understand the writer’s emotional state. By analysing emojis, NLP can offer insights into the writer’s emotions behind a journal entry. Additionally, a summary of emojis at the end of a journal entry has the potential to motivate further journaling. To achieve that we can, for example, use EmoTag1200 dataset[2] which comprises human ratings that associate 150 commonly used emojis with 8 distinct emotions.
Multimodal emotion analysis is another advanced NLP tool that can reason about text while taking into account other modalities like image, audio, and more. Imagine uploading pictures to your journal entry, and an app analyses not only the text but also images to understand your emotions better. Computer vision, a branch of artificial intelligence (AI), empowers computers and systems to extract valuable insights from digital images, videos, and other visual data, and utilise that information to initiate actions or provide recommendations. For instance, by using image captioning techniques, we can identify objects in a picture and potentially establish links between these objects and the emotions of the user. NLP can also take into account music to analyse and understand emotions more deeply because music both induces and expresses emotion.expresses emotions and music induces emotions.
Recent studies show that NLP can be used to detect mental health issues from text. Zhang et al. (2022) suggest that deep learning -based methods have increased in popularity in the last couple of years, demonstrating the great research value and prospects for automatic mental illness detection from text[3]. By utilising advancements in NLP-driven mental illness detection, therapists can detect early symptoms of mental disorders like depression [1].
Finding a suitable therapist and therapy approach can be challenging. NLP can help match users based on their journal entries to the best therapy type, such as cognitive-behavioural therapy (CBT), psychodynamics, or solution-oriented therapy. This method can help users and therapists find each other more easily, enhancing the quality and effectiveness of therapy sessions.
In conclusion, NLP tools offer a wide range of benefits that can enhance the journaling experience and contribute to maintaining good mental health. As NLP technology continues to advance, the potential applications for improving mental health are unlimited. By leveraging the power of NLP, we can gain valuable insights into our emotions, behaviours, and mental health, leading to greater self-awareness and a more fulfilling life.
By now we have been learning a lot about the power of journaling for mental health. It is through technologies like NLP and the tools introduced above, that we have been able to create a journal that writes you back: maind offers a place to write your journal, while picking those crucial keywords, emotions and even emojis from journal entries. While journaling in itself is an effective tool to encourage reflection, and guided journaling can prompt more positive thoughts, it is through technology that we can detect and guide for the best mental health support.
You can already give it a go and see how it works for yourself. maind is available for download on Android and iPhone!
Meanwhile, we also take this opportunity to introduce our team: the great “mainds” behind the app! You’ve all already met Ines, Giouli, and Kleopatra from their blog posts““Dear diary…” and the journal that writes you back (maind Monday 1/5)”, “Unlocking the power of your pen: The psychological benefits of journaling (maind Monday 2/5)”, and “Guided journaling: Understanding the benefits, types, and tools (maind Monday 3/5)”. This week‘s author is our very own NLP Specialist Sardana Ivanova:
Meet the great m[ai]nds!
Sardana Ivanova, PhD candidate and NLP Specialist at Kasvu Labs. Sardana is our NLP specialist from Siberia! She is Sakha herself, and she is writing a PhD thesis about language technology tools for low-resource languages. Also, she works on development of various language technology tools for Sakha. Loves sloths, especially the 3-toed ones, and relates to sloths because she likes a relaxed way of living.
References
- Al-Mosaiwi, M., & Johnstone, T. (2018). In an Absolute State: Elevated Use of Absolutist Words Is a Marker Specific to Anxiety, Depression, and Suicidal Ideation. Clinical Psychological Science, 6(4), 529–542. https://doi.org/10.1177/2167702617747074
- Abu Awal Md Shoeb and Gerard de Melo. 2020. EmoTag1200: Understanding the Association between Emojis and Emotions. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 8957–8967, Online. Association for Computational Linguistics.
- Zhang, T., Schoene, A.M., Ji, S. et al. Natural language processing applied to mental illness detection: a narrative review. npj Digit. Med. 5, 46 (2022). https://doi.org/10.1038/s41746-022-00589-7
