animal shelter¶

  • The dataset contatins information about animals that are adopted from Bloomington Animal Shelter
  • Based on previous studies, animals that are shelterd in an adoption site is more likely to be adopted based on how long they have been at the site and their phenotype (breed).
  • Size and reasons why the dog is in shelter also plays a significant role.

  • Based on this csv data set about animals in Bloomington Animal Shelter, I would like to learn more about specifically what type of dog is more likely to be adopted based on many different features.

  • Citations Citation 1 Citation 2 Citation 3

In [ ]:
In [9]:
import pandas as pd

'''animal shelter data set

Attributes:

    intakedate: Date of admission to the shelter. (Date)
    intakereason: Reason for admission to the shelter. (String)
    istransfer: Whether the animal was transferred from another shelter. (Boolean)
    sheltercode: Unique code assigned to the animal. (String)
    animalname: Name of the animal. (String)
    breedname: Breed of the animal. (String)
    basecolour: Color of the animal. (String)
    speciesname: Species of the animal. (String)
    animalage: Age of the animal. (Integer)
    sexname: Sex of the animal. (String)
    location: Location of the animal. (String)
    movementdate: Date of movement. (Date)
    movementtype: Type of movement. (String)
    istrial: Whether the animal was on trial. (Boolean)
    returndate: Date of return. (Date)
    returnedreason: Reason for return. (String)
    deceaseddate: Date of death. (Date)
    deceasedreason: Reason for death. (String)
    diedoffshelter: Whether the animal died offsite. (Boolean)
    puttosleep: Whether the animal was euthanized. (Boolean)
'''

# Load the CSV file 
df = pd.read_csv('animal-data-1.csv')

# View the first few data dictionary 
(df.head())
Out[9]:
index id intakedate intakereason istransfer sheltercode identichipnumber animalname breedname basecolour ... movementdate movementtype istrial returndate returnedreason deceaseddate deceasedreason diedoffshelter puttosleep isdoa
0 0 15801 2009-11-28 00:00:00 Moving 0 C09115463 0A115D7358 Jadzia Domestic Short Hair Tortie ... 2017-05-13 00:00:00 Adoption 0.0 NaN Stray NaN Died in care 0 0 0
1 1 15932 2009-12-08 00:00:00 Moving 0 D09125594 0A11675477 Gonzo German Shepherd Dog/Mix Tan ... 2017-04-24 00:00:00 Adoption 0.0 NaN Stray NaN Died in care 0 0 0
2 2 28859 2012-08-10 00:00:00 Abandoned 0 D12082309 0A13253C7B Maggie Shep Mix/Siberian Husky Various ... 2017-04-15 00:00:00 Adoption 0.0 NaN Stray NaN Died in care 0 0 0
3 3 30812 2013-01-11 00:00:00 Abandoned 0 C1301091 0A13403D4D Pretty Girl Domestic Short Hair Dilute tortoiseshell ... 2017-04-18 00:00:00 Foster 0.0 2018-05-29 00:00:00 Stray NaN Died in care 0 0 0
4 4 30812 2013-01-11 00:00:00 Abandoned 0 C1301091 0A13403D4D Pretty Girl Domestic Short Hair Dilute tortoiseshell ... 2018-05-29 00:00:00 Adoption 0.0 NaN Stray NaN Died in care 0 0 0

5 rows × 24 columns

How will the data be used to solve the problem?¶

I will mainly be using the key attributes of animalage, sexname, speciesname, basecolour, breedname, istransfer, intakereason, and intakedate for machine learning methods to test if there are any natural groupings of attributes that are more likely to be adopted. Based on this number, I will make visualization graphs to help picture based on certain attributes like how much certain age, species, reasons for admitting, etc has affected certain animals to be more likely to be adopted than the other or which one is more likely to not be adopted and stay in animal shelter. I think by studying mean, median, of certain traits could definelty help make evidence for hypothesis on why certain traits are more likley to be adopted.

In [ ]: