# Meeting Notes > 4.8.2020, meeting with gilad discussing psychological aspects of the project. ### Labeling in the learning stage * gilad suggested using the PANAS questionare, an established method for messuring the emotinal state of people in a given moment. * this method has a downside of being quite long and requiring a user to answer the questionare too much might skew the results. * a second approach is to use machine learning methods in order to label the data * the first method is to use facial regonition to find out how the user is feeling. * the second arguably more accurate method is to use NLP to analyze the text the user types. The simplest approach would be to use NRC, a simple to use sentiment analysis model. ### Emotions to focus on Gilad suggested to stick to the emotions that were already present in the shortened PANAS test (20 emotions) or the ones found in the NRC model. *we would like to focus our emotinal range as much as possible in order to achive more acurate results.* We also discussed yuval shahars suggestion to find out what the users "stable" emotions looks like, but gilad said we shoud not look for a stable emotion as people dont have them, they are always leaning to one emotion or they other even when not continuously aware of it, hence we should not try and find a stable emotion and focus on gathering good data for the emotional state of the user. ### Project goals We discussed a few possible end goals for the projects, all need to be inspeceted further in order to decide what direction to take. * improving the users mood: we might use the system to detect when the user is in a bad mood, and display and image, text or link to improve the mood. gilad added that there are *counter acts* for different emotions we could use. * a tool for psychologists to monitor and help patients: we could "market" the project as a tool to help psychologists monitor the emotional state of patients throughout the day without needing to invade the patients privacy. * gilad is hesitant about this aprach and thinks it wont be a helpful tool for psychologists. * gilad added that there is market intrest in such projects and mentioned **joy ventures** an investment firm intrested in such technoligies. The project can go to the commercial aspect because it has needs in the market. ## Tasks - [ ] send email to gilad about the name of the psychologist that works with personalized feelings and changing them. - [ ] send email to gilad about the name of the company that tried to get emotions from writing style (he advised them). - [ ] need to update gilad how we're going to gather the data and features.