圖書館異質資源整合檢索系統之使用者介面評估研究

Abstract

近年來圖書館之電子資源館藏快速成長，面對龐大多元的電子資源，使用者在搜尋時，常不知如何選擇合適的資料庫，且對每種資料庫的內容與使用不夠熟悉，因此搜尋時常感無所適從。近來不少圖書館引進異質資源整合檢索系統（簡稱Metasearch），即是希望便利使用者透過單一檢索介面，達到一次查全的目的。

這類系統發展至今，相關研究多偏重於技術應用探討，對於系統之使用性與使用者實際使用情形欠缺瞭解。因此希望以使用者角度，瞭解使用者對Metasearch的認知情況與系統使用性。本研究以目前臺灣地區最多大學圖書館採用的MuseSearch為研究平台。邀請學界與產業界熟悉Metasearch的介面設計專家、學者與不同學科背景之研究生為對象。研究設計分兩階段。首先收集介面設計原則之相關文獻，歸納重要的設計元素。再由專家學者評估各項介面設計原則應用於Metasearch 之適用性與重要性，以產生適當的Metasearch介面評估準則。第二階段藉由一般受試者根據個人研究需求自訂檢索任務，共進行兩次測試，避免受試者因不熟悉系統介面影響評估結果。本研究使用電腦螢幕錄製軟體記錄受試者之檢索歷程與放聲思考內容，輔以觀察法，蒐集使用情況。最後綜合檢索歷程分析結果、介面評估問卷與訪談內容，分析Metasearch之使用性。

研究結果發現，受試者普遍認同Metasearch一次查全的便利性。但對於介面設計與檢索功能，滿意度較低，整體而言使用效率與效益尚未彰顯。根據專家受試者與一般受試者對於Metasearch之評估結果，介面評估專家、資訊檢索專家與一般受試者之認知較為一致。僅對系統資訊之表達方式是否簡單明確，彼此認知差異較大。系統管理者與一般受試者對於介面之簡單明確、功能彈性、操作自由度、資訊易視性、系統整體使用性之認知差異較大。主要原因來自一般受試者重視檢索結果的排序與組織，希望系統自動集中重複的結果，呈現相關排序並給予適當的群集分類，以便利篩選與判斷。而系統管理者考量系統需要足夠的時間，檢索數量龐大的異質資源。為縮短使用者等待時間，檢索結果依照資料取回順序呈現。兩者考量的層面不同，造成受試者滿意度不佳。這方面，雖有某些系統進行改善，但使用者等待檢索結果出現的時間相對加長，檢索範圍亦有限制。整體而言，Metasearch之使用性仍有改善空間，技術瓶頸有待突破，若能即時整合檢索結果，並提供適當的呈現方式，將有助於Metasearch之接受度。In recent years, the electric resource in the library has increased rapidly, when encounter with the abundant and diverse electric resource, the uses often have difficulties to choose proper databases, and are unfamiliar with the contents and utilizations of each database, therefore feel indecisive. Recently many libraries introduced Metasearch system, expect to make the users feel more convenient to make a single search across heterogeneous multiple resources and get what they need.

The developments of the kind of system are mostly relevant with the discussion of technological applications, however lacking of understanding of the usability and the actual using circumstances of the system. Therefore, expect to identify the cognizance of the users toward the system and the usability of it from the angle of the users. This study applied the most popular Metasearch system in the academic library in Taiwan as the study platform. Invite the interface designers, scholars, and graduate students from with different background from the academic circle and industries as the research object. The design of this study can be divided into two phases. First, collect the references of user interface design principles, generalize important design elements, and then evaluated by the professionals of the suitability and importance of each interface design principle applies on Metasearch, in order to generate the proper evaluation criterion of Metasearch interface. The second phases proceed with two examinations on the participants who set up for search tasks according to their personal research demands, in order to avoid the influence from the unfamiliarity of the system. The study monitored the participants searching experiences and oral expressed thoughts with computer recording software, and assist with observations to collect the using condition. Finally, generalize the search log analysis, interface evaluation questionnaires, and interviews to analyze the usability of Metasearch.

The results showed that the participants commonly recognized the convenience of single search across heterogeneous multiple resources of Metasearch. However they showed less satisfaction on the interface design and the search function. Overall speaking, the using efficiency and effectiveness were not shown forth. According to the evaluation from professional participants and general participants of Metasearch, the interface evaluating specialist, information retrieval professionals and general participants have more agreements, they only showed diverse opinions on whether there are simplicity and clarity of system expression or not. The system managers have different cognitions on the simplicity, flexibility, freedom, visibility, and usability of the system. The main reason is because that general participants emphasize the sorting and organizing of the search results, expect the system to automatic deduping, present relevance ranking, and provide proper clustering in order to benefit the selection and determination. However, the system managers consider that the system requires sufficient time to retrieve the great amount of heterogeneous resources. Therefore, in order to shorten the waiting time, the search results should show in the order of the retrieved time. The different considerations cause the low satisfaction of the participants. As for this aspect, although some systems have proceeded with improvements, however the waiting time is relatively increased. There are still technologic difficulties that required to be overcome, if enable to integrate the search results at once, and provide proper display methods will assist the acceptability of Metasearch.