Aditya, Erindra Reynaldi Diaz (2023) Optimalisasi Strategi Pemasaran Bank dengan Analisis dan Visualisasi Menggunakan LightGBM dan SHAP. Undergraduate thesis, UPN Veteran Jawa Timur.
|
Text (Cover)
19082010119-cover.pdf Download (9MB) | Preview |
|
|
Text (Bab 1)
19082010119-bab1.pdf Download (35kB) | Preview |
|
![[img]](http://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Bab 2)
19082010119-bab2.pdf Restricted to Registered users only until 26 July 2025. Download (326kB) | Request a copy |
|
|
Text (Bab 3)
19082010119-bab3.pdf Restricted to Registered users only until 26 July 2025. Download (44kB) | Request a copy |
|
|
Text (Bab 4)
19082010119-bab4.pdf Restricted to Registered users only until 26 July 2025. Download (1MB) | Request a copy |
|
|
Text (Bab 5)
19082010119-bab5.pdf Download (23kB) | Preview |
|
|
Text (Daftar pustaka)
19082010119-daftarpustaka.pdf Download (349kB) | Preview |
Abstract
The marketing campaign in a bank is one of the ways for a bank to achieve its organizational goals, and optimal marketing is a crucial factor for the bank's success in attracting and retaining customers. In this case study, it can be observed that the number of customers subscribing to time deposits is relatively smaller, with 5,289 customers making deposits and 5,873 customers not making deposits. Therefore, if the bank's marketing campaign is conducted suboptimally, it will be challenging to achieve the goals of the campaign. This study aims to optimize the bank's marketing strategy by applying analysis using the LightGBM algorithm, which is a highly effective and efficient Gradient Boosting Decision Tree (GBDT) algorithm, along with the SHAP (Shapley Additive Explanations) visualization technique to facilitate the understanding of the analysis conducted by LightGBM. This will help in designing a more optimal marketing strategy. The interpretation using SHAP, in the form of visualization, reveals that the most influential and relevant factors in the success of the marketing campaign in the Portuguese bank are attributes such as duration, contact_cellular, poutcome_success, month_jun, day, housing, pdays, age, balance, campaign, and education_tertiary. The accuracy score of the resulting predictive model is 0.8584, with an F1 score of 0.8564, and a total of 974 true negatives and 943 true positives.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA76.6 Computer Programming | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Information Systems | ||||||||||||
| Depositing User: | Erindra Reynaldi Diaz Aditya | ||||||||||||
| Date Deposited: | 27 Jul 2023 01:33 | ||||||||||||
| Last Modified: | 27 Jul 2023 01:33 | ||||||||||||
| URI: | http://repository.upnjatim.ac.id/id/eprint/16499 |
Actions (login required)
 |
View Item |