香蕉品質分析

Abstract

本研究的目的在於透過評估不同品種、種植地區等因素對香蕉品質的影響，找出影響香蕉品質的關鍵變數，以協助提高香蕉產業的競爭力和經濟效益。研究過程中，使用了來自Kaggle平台的香蕉品質資料集，並採用多元迴歸分析方法來探討各解釋變數對香蕉品質評分的影響。其中，資料集包括不同品種和地區的香蕉樣本，並涵蓋成熟度、糖分含量、堅硬度、長度、重量等變數。研究結果顯示，糖分含量、成熟度指數、堅硬度和長度對香蕉品質評分具有顯著影響。進一步的ANOVA分析也表示，這些變數能有效地解釋品質評分的變異，且調整後的模型具有較高的解釋力。經過模型調整後，發現台灣適宜種植的香蕉品種為Lady Finger、Blue Java和Burro，而對於農民而言，可根據這些結果來監控香蕉的糖分含量、長度和成熟度，以提高生產效率。本研究的結論可應用於香蕉產業中，協助農民選擇適合的品種及管理措施，進一步提升香蕉品質與市場競爭力。

The objective of this study is to identify the key variables affecting banana quality by evaluating the effects of different varieties and growing regions on banana quality, so as to help improve the competitiveness and economic efficiency of the banana industry. In the course of the study, the banana quality dataset from the Kaggle platform was used, and multiple regression analysis was used to explore the influence of various explanatory variables on the banana quality score. The dataset includes banana samples of different varieties and regions, and covers variables such as ripeness, sugar content, firmness, length, weight, etc. The results showed that sugar content, ripeness index, firmness and length had a significant effect on banana quality scores. Further ANOVA analysis also showed that these variables could effectively explain the variation of quality scores, and the adjusted model had high explanatory power. After adjusting the model, it was found that the suitable banana varieties in Taiwan were Lady Finger, Blue Java, and Burro, and for farmers, these results could be used to monitor the sugar content, length, and ripeness of bananas to improve production efficiency. The conclusions of this study can be applied to the banana industry to help farmers choose suitable varieties and management measures to further improve the quality and market competitiveness of bananas.