NIRVATECHS Psychological Analyzer & Personalities types Classifier Slide 1 300 million people have been diagnosed with depression 800,000 people die due to suicide every year How do we make decisions? Emotional State Personality Type What is NIRVATECHs? Emotion Detection Personality Type Classification Pre-processing & Features engineering Research & Data collection Nirvatechs Application Building Stages Building models & Training 1 2 3 4 Research & Data Collection Emotion detection datasets - labeled Here we should put data samples example Emotion detection datasets - unlabeled Emotion detection datasets - validation Average vector of more than one labeling record 1 AWS Mechanical Turks 2 Personality type datasets - labeled Emotion detection datasets - Results Emotion Detection 416,000 Labeled record Personality Types 10,000,000 Labeled record Pre-processing & features engineering Punctuation marks removal Numbers & symbols elimination Spelling error correction Stop words removal 1 2 3 4 Pre-processing & features engineering # words # Links # Videos 50 3 2 Pre-processing & features engineering Feature description Feature Description CC coordinating conjunction RB adverb CD cardinal digit VB verb DT determiner COUNT_W words per sequence EX existential there like: “there is” COUNT_L links per sequence FW foreign word COUNT_P photos per sequence IN preposition COUNT_V videos per sequence JJ adjective ‘big’     MD modal ‘could’ NN noun PRP personal pronoun Building Models & Training Building Models - structure Building Models - structure Introversion (I) – Extroversion (E) Intuition (N) – Sensing (S) Thinking (T) – Feeling (F) Judging (J) – Perceiving (P) Building Models & Training Emotion detection – 94% for >150 words text sequence MBTI classifier – 79% for > 2000 words text sequence Nirvatechs mobile Application Where are We? 1 2 3 4 5 Web platform Collecting data AWS Mechanical turks First model Emotion detection Second model MBTI analysis The application Alpha version Future work Future work Future work Slide 1 THANK YOU Competitive advantage The use of Deep Machine Learning Others Boring, very long questionnaires Others Culture and language are not very well considered Others Services takes days/weeks to be served Others Storing data about users, and do it in traditional way or non-smart websites Locality & Customization Of Language Instant. accurate Services Secure. Private. Virtual. Sales & marketing 28 NIRVATECHS How we will make money? Selling services as packages: Monitoring and Tracking psychological state: 4$/Month Live Sessions with Therapists: 70$/Month Personal Assistant service:10$/Hr Parent Account subscription: 1$/Month Advertisements: Customized Ads depends on the psychological analysis of users, price is based on the area and type of ads. Affiliate Marketing: Commission on products offered to users based on interests of users. How we will reach customers? Online Advertisements. Local Booths and events. Partnerships with Psychiatric Clinics. competitors Your Logo Here 29 Better Help: Employees 318(78% female) from 19 to 72 Founded 2013 Rev $10.5m/year Rev in 2018 reach $ 60 m Users 1 million 2500 us professionals 35-70 $ per week 7 Cups: Employees (est.) (Mar 2019) 210 (+4%) Website Visits (Jan 2019) 1.7 m total Funding $59 m Categories Health Care, Information Technology, Psychology, Wellness Headquarters Regions San Francisco Bay Area, Silicon Valley, West Coast Founded Date 2013 Founders Glen Moriarty Operating Status Active Rev: 7m$/year competitors Your Logo Here 30 Talk Space: Employees (est.) (Mar 2019) 466 (+3%) Website Visits (Jan 2019) 1.1 m total Funding $59 m Categories Employee Benefits, Health Care, Wellness Headquarters Regions San Francisco Bay Area, West Coast, Western US Founded Date 2017 Founders Alyson Friedensohn, Erica T. Johnson Operating Status Active rev: $5.4M/year Ginger.io: Employees (est.) (Mar 2019) 105 (+10%) Website Visits (Jan 2019) 20.7 k total Funding $28.2m Categories Analytics, Big Data, Health Care, mHealth, Mobile Headquarters Regions San Francisco Bay Area, West Coast, Western US Founded Date 2011 Founders Alex Pentland, Anmol Madan, Karan Singh, Ryan O'Toole Operating Status Active rev: $4M/year Business model NIRVATECHS 31 Social Media Sources Google Other DATA Providers Value Proposition monitoring the psychological states decisions recommendation system psychological behavioural enhancer live sessions for human therapists Proposition system by mood and interests Hot-Line consultation system Customer Segments students (smart devices holders) school students, high school students, university students Adults (aged 22+) Parents Parents who have children -18 years old Our Employees Revenue Streams Advertisements App's features and services packages In-App offers Affiliate Marketing Sell smart watches & rings Personal assistants commission Business model NIRVATECHS 32 Key Resources Developers, testers, UX designers, etc. Databases and data providers human resources psychologists, therapists, translators, etc.. Cost Structure Research and Development Licensed software Marketing expenses Hardware resources People developing and maintaining software Key Partners Social Media Sources Google Other DATA Providers Hosted building Around 800,000 person die due to suicide every year ! NIRVATECHS 33 WHO global health estimates – 2016 male 10 to 14 15 to 19 20 to 24 25 to 29 10 to 29 5130 25690 45646 51074 127540 female 10 to 14 15 to 19 20 to 24 25 to 29 10 to 29 4238 27060 33046 30209 94553 both 10 to 14 15 to 19 20 to 24 25 to 29 10 to 29 9368 52750 78691 81284 222093 Statistics references Your Logo Here 34 300 million people around the world have depression Depression and Other Common Mental Disorders: Global Health Estimates. Geneva: World Health Organization; 2017. Around 800,000 person die due to suicide every year ! WHO global health estimates – 2016 Only 1 in 5 people receive treatment consistent with current practice guidelines Depression and Other Common Mental Disorders: Global Health Estimates. Geneva: World Health Organization; 2017. 74% of People think that we need an APP for psychological health care and analysis. Our published Questionnaire image1.png image2.png image3.png image4.png image5.png image6.png image7.png image16.png image17.png image8.png image9.png image10.png image11.png image12.png image13.png image14.png image15.png image18.png image19.png image20.png image21.png image22.png image23.PNG image24.PNG image25.png image26.PNG image27.png image28.png image29.jpeg image30.png image31.jpeg image32.png image33.jpeg image34.png image35.jpeg image36.jpeg image37.jpeg image38.png image39.png image40.png image41.png image42.png image43.png image44.png image45.png image46.png image47.png image48.png image49.jpeg image50.png image51.png image52.png image53.png image54.png