Nothing Special   »   [go: up one dir, main page]
Scribd Logo
Upload

Welcome to Scribd!

  • 33 views

    CLUSTER ANALYSIS Devashish

    Uploaded by

    Raman Kulkarni
    The document analyzes customer data using cluster analysis to group customers into profiles. It finds that there are two distinct clusters/profiles. Profile 1 consists of more modern customers who are open to new trends and brands and see personal appearance as a reflection of themselves. Profile 2 consists of older, more traditional customers who prefer established brands and are less open to new trends.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd

    CLUSTER ANALYSIS Devashish

    Uploaded by

    Raman Kulkarni
    33 views4 pages
    The document analyzes customer data using cluster analysis to group customers into profiles. It finds that there are two distinct clusters/profiles. Profile 1 consists of more modern customers who are open to new trends and brands and see personal appearance as a reflection of themselves. Profile 2 consists of older, more traditional customers who prefer established brands and are less open to new trends.

    Original Description:

    Original Title

    CLUSTER ANALYSIS devashish.docx

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document analyzes customer data using cluster analysis to group customers into profiles. It finds that there are two distinct clusters/profiles. Profile 1 consists of more modern customers who are open to new trends and brands and see personal appearance as a reflection of themselves. Profile 2 consists of older, more traditional customers who prefer established brands and are less open to new trends.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    33 views4 pages

    CLUSTER ANALYSIS Devashish

    Uploaded by

    Raman Kulkarni
    The document analyzes customer data using cluster analysis to group customers into profiles. It finds that there are two distinct clusters/profiles. Profile 1 consists of more modern customers who are open to new trends and brands and see personal appearance as a reflection of themselves. Profile 2 consists of older, more traditional customers who prefer established brands and are less open to new trends.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    of 4

    NAME- DEVASHISH CHATTERJEE, MANISHA MASUD

    PGDM ID- 379, 2113


    ASSIGNMENT- M2 / CLUSTER ANALYSIS

    OBJECTIVE- To find the no of cluster and give profile to them.


    Agglomeration Schedule

    Stage Cluster First


    Cluster Combined Appears

    Stage Cluster 1 Cluster 2 Coefficients Cluster 1 Cluster 2 Next Stage


    1 23 40 .000 0 0 31
    2 22 39 .000 0 0 31
    3 21 38 .000 0 0 29
    4 20 37 .000 0 0 37
    5 19 36 .000 0 0 23
    6 18 35 .000 0 0 32
    7 17 34 .000 0 0 28
    8 6 33 .000 0 0 24
    9 5 32 .000 0 0 21
    10 4 31 .000 0 0 21
    11 10 30 .000 0 0 26
    12 9 29 .000 0 0 27
    13 8 28 .000 0 0 39
    14 7 27 .000 0 0 23
    15 3 26 .000 0 0 25
    16 2 25 .000 0 0 22
    17 1 24 .000 0 0 27
    18 13 16 .000 0 0 22
    19 12 15 .000 0 0 37
    20 11 14 .000 0 0 26
    21 4 5 8.000 10 9 25
    22 2 13 9.000 16 18 24
    23 7 19 11.000 14 5 30
    24 2 6 11.500 22 8 30
    25 3 4 14.000 15 21 28
    26 10 11 15.000 11 20 33
    27 1 9 15.000 17 12 29
    28 3 17 16.667 25 7 32
    29 1 21 19.500 27 3 36
    30 2 7 20.167 24 23 34
    31 22 23 21.000 2 1 33
    32 3 18 22.000 28 6 34
    33 10 22 23.500 26 31 35
    34 2 3 25.880 30 32 35
    35 2 10 30.450 34 33 36
    36 1 2 33.833 29 35 38
    37 12 20 34.000 19 4 38
    38 1 12 41.294 36 37 39
    39 1 8 49.000 38 13 0

    JUSTIFICATION- Since all the value are in interval scale so we will use clustering method to
    fulfill our objective.

    INFERENCE-
    Coefficient Stage Cluster First Next
    Stage Cluster Combined s Appears Stage
    Cluster 1 Cluster 2 Cluster 1 Cluster 2
    37 12 20 34 19 4 38 34
    41.294117 7.2941176
    38 1 12 6 36 37 39 5
    7.7058823
    39 1 8 49 38 13 0 5

    Since the difference between clusters are decreasing, so we can infer that there are 2 clusters
    that can be formed.

    ANOVA

    Cluster Error
    Mean Square df Mean Square df F Sig.
    dont buy product not from
    an established brand 23.645 1 1.301 38 18.169 .000
    buy only new products tried
    & tested as safe 5.682 1 1.627 38 3.493 .069
    I know most cosmetic
    brands 7.471 1 1.288 38 5.802 .021
    one company cannot
    provide complete 2.945 1 1.512 38 1.948 .171
    personalcare solution
    plan shopping trips
    carefully 2.231 1 1.233 38 1.809 .187
    personal care companies
    need to do a lot of research 19.236 1 1.273 38 15.114 .000

    important to look good and


    presentable in todays times 5.531 1 1.128 38 4.903 .033

    like experimenting with new


    trends and style 10.925 1 1.460 38 7.484 .009
    go by what filmstars
    endorse 15.531 1 1.128 38 13.767 .001
    what i wear reflects who I
    am 31.289 1 1.029 38 30.400 .000

    The F tests should be used only for descriptive purposes because the clusters have been chosen to
    maximize the differences among cases in different clusters. The observed significance levels are not
    corrected for this and thus cannot be interpreted as tests of the hypothesis that the cluster means are
    equal.

    dont buy product not from an established brand Sig


    I know most cosmetic brands Sig
    personal care companies need to do a lot of sig
    research
    important to look good and presentable in Sig
    todays times
    like experimenting with new trends and style Sig
    go by what filmstars endorse Sig
    what i wear reflects who I am sig

    Final Cluster Centers

    Cluster
    1 2
    dont buy product not from
    an established brand 3.55 2.00
    buy only new products tried
    & tested as safe 3.09 2.33
    I know most cosmetic
    brands 2.91 3.78
    one company cannot
    provide complete 2.45 3.00
    personalcare solution
    plan shopping trips
    carefully 2.64 3.11
    personal care companies
    need to do a lot of research 3.73 2.33

    important to look good and


    presentable in todays times 2.36 3.11

    like experimenting with new


    trends and style 2.73 3.78
    go by what filmstars
    endorse 4.36 3.11
    what i wear reflects who I
    am 2.00 3.78

    VARIBLES CLUSTER 1 CLUSTER 2 PROFILE 1 PROFILE 2


    dont buy product NEG POS Modern Old
    not from an customer customer
    established brand
    I know most POS NEG
    cosmetic brands
    personal care POS NEG
    companies need to
    do a lot of research
    important to look POS NEG
    good and
    presentable in
    todays times
    like experimenting pos Neg
    with new trends and
    style
    go by what filmstars neg Neg
    endorse
    what i wear reflects pos neg
    who I am

    You might also like