FREE ELECTRONIC LIBRARY - Dissertations, online materials

Pages:     | 1 |   ...   | 2 | 3 ||

«Keyword-based searching and clustering of news articles have been widely used for news analysis. However, news articles usually have other attributes ...»

-- [ Page 4 ] --

After they finished all three tasks, we also asked the users about their general feeling about our design. The biggest concern we had before conducting the user study was the visual clutter and distraction in the interface. However, the responses to our system from users are quite enthusiastic. Overall, they feel the system is informative, intuitive, and visually appealing. In particular, they think that the visual interface is well organized, and each of the visual component has a very clear purpose, which makes the exploration easy.

We also asked their opinions for each visual component. Most of them appreciate the familiar metaphors, such as wheels, belts, and chains, to intuitively represent document streams and sophisticated relations involving documents, keywords, and sentiment values. Among all the visual elements, the users appreciate the wheel most, because it combines multiple aspects of information in a natural and intuitive way and they do not need to think too much to understand the meanings during the fast exploration. In addition to the wheel, they also agree that the significance curve is a very important feature, because without the overview, they will have no choice but to wildly explore the whole document stream, which could make them feel insecure. For example, during the free exploration, they were all interested in those time points when the curve is very steep, and spent some time there trying to figure out why. Therefore, they also feel that the accuracy of the curve is extremely important. Otherwise, the curve is misleading.

In addition to those positive comments, they also raised a few constructive suggestions. For example, five users feel that using thickness to perform quantitative comparison is not very efficient. Therefore, they suggested we provide an option to show actual numbers when the thicknesses are too similar to compare. Another suggestion is regarding the system control. For example, some of the users would like to control the sensitivity of the wheel to the summarized keyword sentiments, so they can ignore those minor fluctuations and focus more on big patterns. Some of them would like to be able to add or remove keywords during exploration other than choosing them before exploring. We think these suggestions are all very valuable, and intended to explore them in future work. Furthermore, we also plan to deploy to a news Web site to reach more audiences and further improve our system based on their comments.

9. DISCUSSIONS From the experiments, we can see that our system has advantages and can encode a lot of information into one display for analysis such that some unexpected correlations may emerge. Meanwhile, we also identify some weaknesses of our system. Our system can handle thousands of documents and tens of keywords effectively. If the news

–  –  –

streams contain too many articles, exploration may still take a very long time and reduce its effectiveness. Too many keywords may overwhelm the keyword wheels and cause clutter in the display. To deal with this problem, it is better to use our system together with some data mining techniques to first narrow down the document scope.

Some well-established techniques in the visualization field, such as clutter reduction methods, can also be applied. As our system provides useful information into one display, it is possible that users are overwhelmed and lose their focus. Thus, we recommend that users turn off some features of the system and only focus on one feature at the beginning. After getting familiar with the data, more details can be brought into the display.

10. CONCLUSION In this article, we have presented a visual analytics system for large-scale news streams. Our system aims at providing the multiple attributes of news articles and keywords, the dynamic relations between news articles and keywords, the micro relation among keywords, and the macro relation among documents simultaneously to users for analysis. We designed an original TextWheel which consists of a document transportation belt, one or multiple keyword wheels, and a chain system to connect the belt and wheel. Our system is based on some everyday objects which users are familiar with and thus the learning curve for our system should be low. We demonstrated the effectiveness of our system by applying it to several news corpora related to some major companies, obtaining some interesting findings. The application of our system is not limited to news streams. It can be used to analyze other data in text format (e.g., emails, blogs, and internal memos), and to reveal the macro/micro relations existing in other data formats, such as video clips.

In the future, we plan to further extend our system to encode more attributes of text documents. We believe the integration of our system with other data mining methods will make it more powerful. We also want to encode the uncertainty associated with the sentiment and co-occurrence computations. Although we conducted a user study which has suggested that our subjects agree that our design and system can help them with pattern huntings in macro/micro relations, it is still preliminary and informal.

Therefore, we also plan to conduct a more thorough user study, which involves more subjects, comparison with other tools, and a formal questionnaire-based survey.

REFERENCES ALBRECHT-BUEHLER, C., WATSON, B., AND SHAMMA, D. 2005. Visualizing live text streams using motion and temporal pooling. IEEE Comput. Graph. Appl. 52–59.

ALLAN, J., HARDING, S., FISHER, D., BOLIVAR, A., GUZMAN-LARA, S., AND AMSTUTZ, P. 2005. Taking topic detection from evaluation to practice. In Proceedings of the Hawaii International Conference on System Sciences. 101–101.

BAUTIN, M., WARD, C., PATIL, A., AND SKIENA, S. 2010. Access: News and blog analysis for the social sciences. In Proceedings of the 19th International Conference on World Wide Web. ACM, 1229–1232.

BRILL, E. 1994. Some advances in rule-based part of speech tagging. In Proccedings of the 12th National Conference on Artificial Intelligence.

CHIEU, H. AND NG, H. 2002. Named entity recognition: A maximum entropy approach using global information. In Proceedings of the International Conference on Computational Linguistics. 190–196.

COVER, T. M. AND THOMAS, J. A. 2006. Elements of Information Theory 2nd Ed. Wiley-Interscience.

ERTEN, C., HARDING, P., KOBOUROV, S., WAMPLER, K., AND YEE, G. 2004. Exploring the computing literature using temporal graph visualization. In Proceedings of the Conference on Visualization and Data Analysis (VDA).

GLANCE, N. S., HURST, M., AND TOMOKIYO, T. 2004. Blogpulse: Automated trend discovery for weblogs.

In Proceedings of the Workshop on the Weblogging Ecosystem: Aggregation, Analysis and Dynamics.

ACM Transactions on Intelligent Systems and Technology, Vol. 3, No. 2, Article 20, Publication date: February 2012.

Watch the Story Unfold with TextWheel: Visualization of Large-Scale News Streams 20:17 GRANITZER, M., KIENREICH, W., SABOL, V., ANDREWS, K., AND KLIEBER, W. 2004. Evaluating a system for interactive exploration of large, hierarchically structured document repositories. In Proceedings of the IEEE Symposium on Information Visualization. 127–134.

HAVRE, S., HETZLER, B., AND NOWELL, L. 2000. Themeriver: Visualizing theme changes over time. In Proceedings of the IEEE Symposium on Information Visualization. 115–123.

HETZLER, E. G., CROW, V. L., PAYNE, D. A., AND TURNER, A. E. 2005. Turning the bucket of text into a pipe. In Proceedings of the IEEE Symposium on Information Visualization (INFOVIS’05). IEEE Computer Society. 89–94.

LLOYD, L., KECHAGIAS, D., AND SKIENA, S. 2005. Lydia: A system for large-scale news analysis. In Proceedings of the 12th Symposium of String Processing and Information Retrieval (SPIRE’05). Lecture Notes in Computer Science, vol. 3772. Springer, 161–166.

MAO, Y., DILLON, J., AND LEBANON, G. 2007. Sequential document visualization. IEEE Trans. Vis. Comput.

Graph. 13, 6, 1208–1215.

MCDONALD, D. 1993. Internal and external evidence in the identification and semantic categorization of proper names. In Proceedings of the Workshop on Acquisition of Lexical Knowledge from Text. 32–43.

MITCHELL, T. 1997. Machine Learning. McGraw-Hill.

OELKE, D., BAK, P., KEIM, D., LAST, M., AND DANON, G. 2008. Visual evaluation of text features for document summarization and analysis. In Proceedings of the IEEE Symposium on Visual Analytics Science and Technology. 75–82.

PANG, B. AND LEE, L. 2008. Opinion Mining and Sentiment Analysis. Now Publishers.

PAULOVICH, F. V. AND MINGHIM, R. 2008. Hipp: A novel hierarchical point placement strategy and its application to the exploration of document collections. IEEE Trans. Vis. Comput. Graph. 14, 6, 1229–1236.

¨ STASKO, J., G ORG, C., AND LIU, Z. 2008. Jigsaw: Supporting investigative analysis through interactive visualization. Inf. Vis. 7, 2, 118–132.

VAN HAM, F., WATTENBERG, M., AND VIEGAS, F. B. 2009. Mapping text with phrase nets. IEEE Trans. Vis.

Comput. Graph. 15, 6, 1169–1176.

´ WATTENBERG, M. AND VI EGAS, F. B. 2008. The word tree, an interactive visual concordance. IEEE Trans.

Vis. Comput. Graph. 14, 6, 1221–1228.

WISE, J. A. 1999. The ecological approach to text visualization. J. Amer. Soc. Inf. Sci. 50, 13, 1224–1233.

WONG, P. C., WHITNEY, P., AND THOMAS, J. 1999. Visualizing association rules for text mining. In Proceedings of the IEEE Symposium on Information Visualization. 120–123.

WONG, P. C., FOOTE, H., ADAMS, D., COWLEY, W., AND THOMAS, J. 2003. Dynamic visualization of transient data streams. In Proceedings of the IEEE Symposium on Information Visualization. 97–104.

Received July 2010; revised October 2010; accepted October 2010

ACM Transactions on Intelligent Systems and Technology, Vol. 3, No. 2, Article 20, Publication date: February 2012.

Pages:     | 1 |   ...   | 2 | 3 ||

Similar works:

«Stock Repurchases and Liquidity∗ Alexander Hillert† Ernst Maug‡ Stefan Obernberger§ September 15, 2014 Abstract We analyze the two-way relationship between share repurchases and liquidity based on a newly available data set of realized share repurchases in the US, which covers 50,204 repurchase months between 2004 and 2010. Repurchases unequivocally improve liquidity. Firms also adapt their buyback programs by reducing repurchase activity when the market for their stock is less liquid....»

«1 PV 5311 AMP ® STRATTERA (atomoxetine HCl) WARNING Suicidal Ideation in Children and Adolescents — STRATTERA (atomoxetine) increased the risk of suicidal ideation in short-term studies in children or adolescents with Attention-Deficit/Hyperactivity Disorder (ADHD). Anyone considering the use of STRATTERA in a child or adolescent must balance this risk with the clinical need. Patients who are started on therapy should be monitored closely for suicidality (suicidal thinking and behavior),...»

«ALTERNATIVE UCITS CONFERENCE SMART AND LIQUID ASSETS FOR PORTFOLIO DIVERSIFICATION METROPOLE HOTEL BRUSSELS 28 JANUARY 2016 • • PROGR AM GOLD PARTNERS SILVER PARTNERS BRONZE PARTNERS LuxHedge does not assume any liability for, nor does it guarantee the completeness, accuracy or up-to-date of the information provided in this document. The description of products and services is provided for information purposes only and does not constitute an offer or investment advice. It cannot be offered...»

«THE DISCOVERY OF LIFE ON MARS Andrew D. Basiago President Mars Anomaly Research Society P.O. Box 2311 Vancouver, WA 98668 U.S.A.ABSTRACT There is life on Mars. Evidence that the Red Planet harbors life and has for eons was discovered by the author by examining NASA photograph PIA10214, a westward view of the West Valley of the Columbia Basin in the Gusev Crater that was taken by the Mars Exploration Rover Spirit in November 2007 and beamed back to the Earth. This photograph has been the subject...»

«S:\Common\Cadets\RAF\Current RAF Admin Mar 2014-\CCF Stuff\Ultilearn\20100825-Ultilearn Student Manual-U-Logs4b.Doc 13 March 2012 Ultilearn User Manual Version 3.0 For The Ultilearn Learning System Headquarters Air Cadets Logistics – Bader (Logs 4) Bader Project Number – 001hb.2 Prepared by: Andrew Bracegirdle Bader Developer HQAC RAF Cranwell NG34 8HB 13 March 2012 1 S:\Common\Cadets\RAF\Current RAF Admin Mar 2014-\CCF Stuff\Ultilearn\20100825-Ultilearn Student Manual-U-Logs4b.Doc...»

«Pan-Pacific Association of Applied Linguistics 14(2), 187-211 Students' Behaviors and Views of Paraphrasing and Inappropriate Textual Borrowing in an EFL Academic Setting Ming-Tzu Liao* and Chiung-Ying Tseng National Kaohsiung Normal University Liao M-T., & Tseng C-Y. (2010). Students’ behaviors and views of paraphrasing and inappropriate textual borrowing in an EFL academic setting. Journal of Pan-Pacific Association of Applied Linguistics, 14(2), 187-211. This study examines skillful EFL...»

«249 Comechingonia. Revista de Arqueología Número 18, segundo semestre de 2014, pp. 249-261, Córdoba ISSN 0326-7911 USO ANTRÓPICO DE LAGARTOS (TUPINAMBIS SP.) EN EL SITIO BELTRÁN ONOFRE BANEGAS-LAMI HERNÁNDEZ (SANTIAGO DEL ESTERO). ANTHROPIC USE OF LIZARDS (TUPINAMBIS SP.) AT BELTRÁN ONOFRE BANEGAS-LAMI HERNÁNDEZ SITE (SANTIAGO DEL ESTERO PROVINCE). Luis M. del Papa1 y Leda Moro2 1 CONICET. Facultad de Ciencias Naturales y Museo, Universidad Nacional de La Plata. Calle 64 s/n, entre 120...»

«(pp. 224–253). Newcastle upon Tyne, England: Cambridge Scholars Publishing. (Translated from 2009 French edition.) Piccardo, E. (2010b). From communicative to action-oriented: New perspectives for a new millennium. CONTACT TESL Ontario, 36(2), 20–35. Piccardo, E., Berchoud, M., Cignatta, T., Mentz, O., & Pamula, M. (2011). Pathways through assessing, learning and teaching in the CEFR. Strasbourg, France: Council of Europe Publishing. Retrieved...»

«Psychological Test and Assessment Modeling, Volume 56, 2014 (1), 25-44 The impact of group pseudo-guessing parameter differences on the detection of uniform and nonuniform DIF W. Holmes Finch1 & Brian F. French Abstract Differential item functioning (DIF) is an important aspect of item development and validity assessment. Traditionally DIF is divided into two broad types, focusing on conditional group differences of the item difficulty (uniform DIF) and discrimination (nonuniform DIF)...»

«www.the-criterion.com The Criterion: An International Journal in English ISSN 0976-8165 From Routes to Roots: Diaspora in the Novels of Salman Rushdie. Janmejay Kumar Tiwari Displacement has no replacement and this is the reality of diaspora. Since the inducement of humanity, people have been straddling throughout the world without having any theoretical knowledge of boundaries. He has always been suffering from the disease of alienation without realizing the actual outcome of migration....»

«Obituaries Sisters of Mercy – South Central Community 2013 Sister Mary Patricia O’Brien, age 92, October 12, 2013 Sister Mary Patricia O’Brien died peacefully following a brief illness at Mercy Hospital in Fort Smith, Arkansas. She was a Sister of Mercy for 74 years. Sister Patricia ministered in nursing administration and nursing education for almost 30 years in Arkansas, Oklahoma, Missouri and Louisiana after graduating from St. John’s School of Nursing in Springfield, Missouri. In...»

«Sketch notes on South Australia's Onkaparinga threshing roller, and some antecedents Bridget Jolly Part 2 More on Kangaroo Island A possibly unique and certainly important threshing floor is near Penneshaw on Kangaroo Island. This was Joseph Henry Frederick (Harry) Bates's (1846-1936) floor. An inventory taken in 1920 of Bates's stock and implements on his 'Ironstone Hill Estate' did not mention a threshing roller, even in the 'Sundries' list.1 If visible at the time, the roller presumably was...»

<<  HOME   |    CONTACTS
2016 www.dissertation.xlibx.info - Dissertations, online materials

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.