The transmission and internalization of values are the primary processes that occur during socialization. A recent approach integrates existing theories and research findings into a comprehensive model of socialization. According to the domains of socialization approach, there is no general principle governing socialization but rather it occurs in different domains of caregiver-child interactions. Grusec and Davidov (2010) outlined five socialization domains, which involve controlling children’s behaviour by external means (control domain), protecting children from harm and relieving their distress (protection domain), teaching children information or skills outside of the discipline or distress setting (guided-learning domain), managing children’s environment to increase desirable role models (group participation domain), and accommodating each other’s wishes (mutual reciprocity domain). Previous work demonstrated the utility of the domains of socialization approach for the study and understanding of value acquisition (Vinik, Johnston, Grusec, & Farrell, 2013). The present study expanded on this work by focusing on processes within the family. A modified narrative methodology was used to explore aspects of the value acquisition process. Autobiographical narratives of 294 emerging adults about a time they learned an important value from a caregiver were analyzed. The sample included participants from four ethnic backgrounds. Findings provided further support for the usefulness of the domains of socialization approach to the study of value development, as events recalled in narratives were categorized into all domains but reciprocity. Values learned in the control domain were most frequently reported but were associated with the lowest levels of internalization. The highest level of value internalization was found to occur in the group participation domain, drawing attention to the importance of observing the behaviour of others. Socialization domains were associated with particular types of lesson content. The guided learning and group participation domains were associated with more positive and less negative emotional valence compared to the other domains. In turn, absence of negative valence was significantly related to better confidence in accuracy of memory reported in narratives, indicative of quality of information processing and learning. Most effects were not moderated by demographic variables providing support to the universal applicability of the domains of socialization approach.

商業知識的發展突飛猛進，其中，預測模型在眾多商業智慧中扮演重要的角色，然而，當我們從大量資料萃取隱藏、未知與潛在具有實用性的資訊處理過程時，往往會遇到許多資料品質上的問題而難以著手分析，尤其是遺失值 (Missing value)的問題在資料前置處理階段更是常見的困難。因此，要如何在建立預測模型時有效的處理遺失值是一個很重要的議題。 過去已有許多文獻致力於遺失值處理的議題，其中，Active Feature-Value Acquisition的相關研究更針對訓練資料的選填順序深入探討。Active Feature-Value Acquisition的概念是從具有遺失值的訓練資料中，選擇適當的遺失資料填補，讓預測的模型在最具效率的情況下達到理想的準確率。本研究將延續Active Feature-Value Acquisition的研究主軸，優先考量決策樹上的節點為遺失值選值填補的順序，提出一個新的訓練資料遺失值的選填順序方法─I Sampling，並透過實際的數據進行訓練與測試，同時我們也與過去文獻所提出的方法進行比較，了解不同的填值偵測與順序的選擇對於一個預測模型的分類準確率是否有影響，並了解各個方法的優缺點與在不同情境下的適用性。 本研究所提出的新方法與驗證的結果，將可給予未來從事預測行為的管理或學術工作一些參考與建議，可以依據不同性質的資料採取合宜的選值方式，以節省取值的成本並提高預測模型的分類能力。 / The importance of business intelligence is accelerated developing nowadays. Especially predictive models play a key role in numerous business intelligence tasks. However, while we extract information from unidentified data, there are critical problems of how to handle the missing values, especially in the data pre-processing phase. Therefore, it is important to identify which methods best deal with the missing data when building predictive models. There are several papers dedicated in the research of strategies to deal with the missing values. The topic of Active-Feature Acquisition (aka. AFA) especially worked on the priority order of choosing which feature-value to acquire. The goal of AFA is to reduce the costs of achieving a desired model accuracy by identifying instances for which obtaining complete information is most informative. Followed by the AFA concept, we present an approach- I Sampling, in which feature-values are selected for acquisition based on the attribute on the top node of the current decision tree. Also we compare our approach with other methods in different situations and data missing patterns. Experimental results demonstrate that our approach can induce accurate models using substantially fewer feature-value acquisitions as compared to alternative policies in some situations. The method we proposed can aid the further predictive works in academic and business area. They can therefore choose the right method based on their needs and obtain the informative data in an efficient way.

預測模型已經被廣泛運用在日常生活中，例如銀行信用評比、消費者行為或是疾病的預測等等。然而不論在建構或使用預測模型的時候，我們都會在訓練資料或是測試資料中遇到遺失值的問題，因而降低預測的表現。面對遺失值有很多種處理方式，刪除、填補、模型建構以及機器學習都是可以使用的方法；除此之外，直接用某個成本去取得遺失值也是一個選擇。 本研究著重的議題是用某成本去取得遺失值，並且利用決策樹(因為其在建構時可以容納遺失值)來當作預測模型，希望可以找到用較低的成本的填值方法達到較高的準確率。我們延續過去Error Sampling中Uncertainty Score的概念與邏輯。提出U-Sampling來判斷不同特徵值的「重要性排序」。相較於過去Error Sampling用「受試者」(row-based)的重要性來排序。U-Sampling是根據「特徵值」(column-based)的重要性來排序。 我們用8組UCI machine Learning Repository的資料進行兩組實驗，分別讓訓練資料以及測試資料含有一定比例的遺失值。再利用U-Sampling、Random Sampling以及過去文獻所提及的Error Sampling作準確率和錯誤減少率的比較。實驗結果顯示在訓練資料有遺失值的情況，U-Sampling在70%以上的檔案表現較佳；而在測試資料有遺失值的情況，U-Sampling則是在87.5%的檔案表現較佳。 另外，我們也研究了對於不同的遺失比例對於上述方法的效果是否有影響，可以用來判斷哪種情況比較適用哪一種選值方法。希望透過U-Sampling，可以先挑選重要的特徵值來填補，用較少的遺失值取得就得到較高的準確率，也因此可以節省處理遺失值的成本。