趨勢資訊內容分析、整合及視覺化模式

The Model for Analysis, Integration and Visualization of Trend Information

黃騰賢(Teng-Hsien Huang)

指導教授 ： 侯建良

國立清華大學/工學院/工業工程與工程管理學系/碩士(2018年)

摘要

當人們欲瞭解一議題之趨勢發展情形（如產業趨勢概況）時，其往往透過網路平台搜尋該標的議題之趨勢的相關報導，以獲得其所關心議題之相關趨勢資訊。然而，於理解所關心議題之相關趨勢資訊的過程中，人們往往需花費時間閱讀各趨勢資訊之內容，並進一步組織同質趨勢資訊中關鍵重點或量化資訊。此外，此些同質趨勢資訊之內容常以文字方式呈現，人們較不易掌握其所對應之趨勢消長狀況而無法有效地進行相關決策。
因此，為協助人們快速掌握感興趣議題之發展趨勢，本研究期望發展一套「趨勢資訊內容分析、整合及視覺化」模式。此模式可先擷取趨勢資訊內容中所包含之趨勢資訊屬性（包含「資訊來源」、「資訊統計時間」、「資訊研究區域」與「資訊調查對象」、「趨勢項目」、「趨勢變量」與「趨勢特徵」等7項屬性），以將趨勢資訊內容結構化；之後，依據趨勢資訊屬性擷取結果，此模式可將同質之趨勢資訊區分至同一趨勢資訊群集；最後，此模式可進一步將各趨勢資訊群集中所有趨勢資訊進行屬性值彙整，並以視覺化方式呈現彙整後之各趨勢資訊群集所對應的趨勢變化情形，以提供資訊需求者明確之趨勢消長狀況。因此，當人們欲瞭解某議題之發展趨勢時，其可透過此模式所提供之視覺化圖表快速且準確掌握該議題之趨勢消長狀況。

關鍵字

分群方法 ； 趨勢分析 ； 趨勢資訊內容整合 ； 趨勢資訊視覺化

並列摘要

As people are interested in the tendency of a specific issue, they usually search for the relevant reports related to the issue through the Internet. They have to collect and organize the contents of these reports to acquire the trend information of the issue that they concern. However, the trend of an issue might exist in several distinct reports. As a result, they have to spend time in browsing and organizing these decentralized trend information. On the other hand, the reports of trend information are usually presented in text form; therefore, it is not easy for readers to efficiently acquire the trend distribution of the issue. Owing to the above problem, this research develops a model for integration and visualization of trend information. The proposed model can be used to integrate and visualize similar trend information. By applying the model, the similar trend information can be clustered and each cluster of trend information can be visualized to reveal the trend distribution.

並列關鍵字

Clustering ； Trend Analysis ； Information Integration ； Information Visualization

參考文獻

1. Alshawi, S., Missi, F. and Eldabi, T., 2003, “Healthcare Information Management: The Integration of Patients’ Data,” Logistics Information Management, Vol. 16, pp. 286-295.

2. Baumgartner, R., Gottlob, G. and Herzog, M., 2009, “Scalable Web Data Extraction for Online Market Intelligence,” Proceedings of the VLDB Endowment, Vol. 2, pp. 1512-1523.

3. Beneki, C., Eeckels, B. and Leon, C., 2012, “Signal Extraction and Forecasting of The UK Tourism Income Time Series: A Singular Spectrum Analysis Approach,” Journal of Forecasting, Vol. 31, pp. 391-400.

4. Cameron, M., Power, R., Robinson, B. and Yin, J., 2012, “Emergency Situation Awareness from Twitter for Crisis Management,” Proceedings of the 21st International Conference on World Wide Web, pp. 695-698.

5. Charbonnier, S. and Portet, F., 2012, “A Self-Tuning Adaptive Trend Extraction Method for Process Monitoring and Diagnosis,” Journal of Process Control, Vol. 22, pp. 1127-1138.