How Our Investment App Recommendation System Works 🚀

1. Collecting User Preferences 📋

Our system gathers user preferences through a series of questions about the 6 main criteria:

  • Ease of Use
  • Educational Features
  • Transaction Costs
  • Security
  • Customer Support
  • Availability of Investment Instruments

For example, here are the preferences selected by a user:

  • Ease of Use: Ease of use is less important to me, as long as the platform offers strong features. (Weight: 0.25)
  • Educational Features: Educational features are somewhat important, but I prefer platforms with low fees. (Weight: 0.5)
  • Transaction Costs: Very low transaction costs are my top priority when choosing a platform. (Weight: 1.0)
  • Security: Security is somewhat important, but I prefer platforms with low transaction costs. (Weight: 0.5)
  • Customer Support: Customer support is very important because I want quick assistance when I encounter issues. (Weight: 1.0)
  • Availability of Investment Instruments: The availability of various investment instruments is very important because I want to have many investment options. (Weight: 1.0)

2. Weight Calculation ⚖️

Each user answer has a specific weight (ranging from 0 to 1.0). This weight is then multiplied by the maximum weight for each criterion to get the actual weight.

Example: If a user selects an option with a weight of 0.75 for the Ease of Use criterion (with a maximum weight of 0.2), the actual weight would be 0.75 * 0.2 = 0.15.

3. Simple Additive Weighting (SAW) Method 🎗️

We use the SAW method to calculate the final score for each application:

  1. Each application starts with an initial score for each criterion (on a scale of 1-9).
  2. These scores are multiplied by the actual weights based on user preferences.
  3. For cost criteria (like Transaction Costs), we use a special formula: (max_score - app_score) / max_score * weight.
  4. All criterion scores are summed up to get the total score for each application.

SAW Calculation Example

Here is the SAW calculation table based on the criteria and scores:

App Ease of Use Educational Features Transaction Costs Security Customer Support Investment Instrument Availability Total Score
Bibit 8 * 0.05 = 0.40 5 * 0.075 = 0.375 (9 - 8) / 9 * 0.2 = 0.022 9 * 0.1 = 0.90 7 * 0.1 = 0.70 6 * 0.15 = 0.90 0.40 + 0.375 + 0.022 + 0.90 + 0.70 + 0.90 = 3.297
Stockbit 9 * 0.05 = 0.45 6 * 0.075 = 0.45 (9 - 6) / 9 * 0.2 = 0.066 7 * 0.1 = 0.7 6 * 0.1 = 0.6 6 * 0.15 = 0.9 0.45 + 0.45 + 0.066 + 0.7 + 0.6 + 0.9 = 3.166
IPOT 5 * 0.05 = 0.25 9 * 0.075 = 0.675 (9 - 5) / 9 * 0.2 = 0.089 8 * 0.1 = 0.8 5 * 0.1 = 0.5 7 * 0.15 = 1.05 0.25 + 0.675 + 0.089 + 0.8 + 0.5 + 1.05 = 3.364
Ajaib 7 * 0.05 = 0.35 8 * 0.075 = 0.6 (9 - 7) / 9 * 0.2 = 0.044 5 * 0.1 = 0.5 8 * 0.1 = 0.8 9 * 0.15 = 1.35 0.35 + 0.6 + 0.044 + 0.5 + 0.8 + 1.35 = 3.644
Bareksa 6 * 0.05 = 0.3 7 * 0.075 = 0.525 (9 - 9) / 9 * 0.2 = 0.022 7 * 0.1 = 0.7 9 * 0.1 = 0.9 9 * 0.15 = 1.35 0.3 + 0.525 + 0.022 + 0.7 + 0.9 + 1.35 = 3.797

4. Score Normalization 🔢

To facilitate comparison, the final scores of each app are normalized:

Here is the table of normalized scores:

App Total Score Normalized Score
Bibit 3.297 3.297 / 3.797 = 0.87
Stockbit 3.166 3.166 / 3.797 = 0.84
IPOT 3.364 3.364 / 3.797 = 0.89
Ajaib 3.644 3.644 / 3.797 = 0.96
Bareksa 3.797 3.797 / 3.797 = 1.00

5. Final Recommendations 📊

Based on the normalized scores, our system recommends:

The system also provides brief reasons why these apps are recommended based on their standout criteria.

Conclusion 📝

Through this process, our system can provide personalized and objective recommendations based on each user's unique preferences, helping them find the investment platform that best meets their needs. This process involves collecting user preferences, calculating weights, applying the SAW method, normalizing scores, and finally providing the most suitable recommendations.