Symptom Assure

Inspiration

Problem which can be solved:

Seven class classifications for each drug separately. Problem can be transformed to binary classification by union of part of classes into one new class. For example, "Never Used", "Used over a Decade Ago" form class "Non-user" and all other classes form class "User". The best binarization of classes for each attribute. Evaluation of risk to be drug consumer for each drug

What it does

Context

Data Set Information:

Database contains records for 1885 respondents. For each respondent 12 attributes are known: Personality measurements which include NEO-FFI-R (neuroticism, extraversion, openness to experience, agreeableness, and conscientiousness), BIS-11 (impulsivity), and ImpSS (sensation seeking), level of education, age, gender, country of residence and ethnicity. All input attributes are originally categorical and are quantified. After quantification values of all input features can be considered as real-valued. In addition, participants were questioned concerning their use of 18 legal and illegal drugs (alcohol, amphetamines, amyl nitrite, benzodiazepine, cannabis, chocolate, cocaine, caffeine, crack, ecstasy, heroin, ketamine, legal highs, LSD, methadone, mushrooms, nicotine and volatile substance abuse and one fictitious drug (Semeron) which was introduced to identify over-claimers. For each drug they have to select one of the answers: never used the drug, used it over a decade ago, or in the last decade, year, month, week, or day.

Detailed description of database and process of data quantification are presented in E. Fehrman, A. K. Muhammad, E. M. Mirkes, V. Egan and A. N. Gorban, "The Five Factor Model of personality and evaluation of drug consumption risk.," arXiv [Web Link], 2015 Paper above solve binary classification problem for all drugs. For most of drugs sensitivity and specificity are greater than 75%

Since all of the features have been quantified into real values please refer to the link to the original dataset to get more clarity on categorical variables. For example, for EScore (extraversion) 9 people scored 55 which corresponds to a quantified (real) value of in the dataset 2.57309. I have also converted some variables back into their categorical values which are included in the drug_consumption.csv file Original Dataset

How we built it

Content

Feature Attributes for Quantified Data:

ID: is a number of records in an original database. Cannot be related to the participant. It can be used for reference only. Age (Real) is the age of participant Gender: Male or Female Education: level of education of participant Country: country of origin of the participant Ethnicity: ethnicity of participant Nscore (Real) is NEO-FFI-R Neuroticism Escore (Real) is NEO-FFI-R Extraversion Oscore (Real) is NEO-FFI-R Openness to experience. Ascore (Real) is NEO-FFI-R Agreeableness. Cscore (Real) is NEO-FFI-R Conscientiousness. Impulsive (Real) is impulsiveness measured by BIS-11 SS (Real) is sensation seeing measured by ImpSS Alcohol: alcohol consumption Amphet: amphetamines consumption Amyl: nitrite consumption Benzos: benzodiazepine consumption Caff: caffeine consumption Cannabis: marijuana consumption Choc: chocolate consumption Coke: cocaine consumption Crack: crack cocaine consumption Ecstasy: ecstasy consumption Heroin: heroin consumption Ketamine: ketamine consumption Legalh: legal highs consumption LSD: LSD consumption Meth: methadone consumption Mushroom: magic mushroom consumption Nicotine: nicotine consumption Semer: class of fictitious drug Semeron consumption (i.e. control) VSA: class of volatile substance abuse consumption Rating's for Drug Use:

CL0 Never Used CL1 Used over a Decade Ago CL2 Used in Last Decade CL3 Used in Last Year 59 CL4 Used in Last Month CL5 Used in Last Week CL6 Used in Last Day

Acknowledgements

Elaine Fehrman, Men's Personality Disorder and National Women's Directorate, Rampton Hospital, Retford, Nottinghamshire, DN22 0PD, UK, Elaine.Fehrman@nottshc.nhs.uk

Vincent Egan, Department of Psychiatry and Applied Psychology, University of Nottingham, Nottingham, NG8 1BB, UK, Vincent.Egan@nottingham.ac.uk

Evgeny M. Mirkes Department of Mathematics, University of Leicester, Leicester, LE1 7RH, UK, em322@le.ac.