MODEL PREDIKSI TRAFIK PEMAKAIAN WIFI (STUDI KASUS PADA WIFI.ID PT TELKOM)
In developing the new product would need a judgment on it. Consideration has been done by PT Telkom currently wifi.id in developing new products is through the level of internet usage is based on two things, namely the use of a home phone number and use flexi smartphones. Considerations based intern...
محفوظ في:
| المؤلفون الرئيسيون: | , |
|---|---|
| التنسيق: | Theses and Dissertations NonPeerReviewed |
| منشور في: |
[Yogyakarta] : Universitas Gadjah Mada
2014
|
| الموضوعات: | |
| الوصول للمادة أونلاين: | https://repository.ugm.ac.id/131872/ http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=72381 |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| الملخص: | In developing the new product would need a judgment on it. Consideration
has been done by PT Telkom currently wifi.id in developing new products is
through the level of internet usage is based on two things, namely the use of a
home phone number and use flexi smartphones. Considerations based internet
usage 2 that, can not predict the use of technology in every province wifi.id. The
problem faced by PT Telkom Internet usage today is far below the target. Making
a prediction model is one way to evaluate the variables that affect the use wifi.id.
Currently, PT Telkom needs a model that can determine the factors that affect the
use of the internet. A study of the factors that may influence and use of predictive
models appropriate to the wifi on a region required in order to reduce the gap
between the target and actual use of the wifi in a province so that predictive
models can be used as one alternative tools for companies in targeting the use of
wifi in a region.
In this study, the object of study is a national wifi.id users. This research will
use the data to use wifi internet usage overall nationally. The study was conducted
by making use of prediction models wifi.id. level The prediction model using the
software Statistical Package for the Social Science 16.0. In this study using three
considerations, namely linear regression model, quadratic, exponential, and a
combination of all three.
From the research conducted, it has been obtained based on the results of the
correlation analysis of nine factors derived factors that significantly affect the rate
of use of wifi is coming of promotional activity, percentage of use of existing wifi
and Gross Domestic Product. With the best prediction model using a combination
of regression R-square value of 84.9% (Cross Validation). |
|---|