大数据挖掘服务研究进展——Studies in Big Data Vol.3

需积分: 9 155 浏览量更新于2024-07-20 收藏 9.3MB PDF 举报

"Studies in Big Data 03" 是一本专著，由 Katsutoshi Yada 担任编辑，收录在Springer出版社的系列“Studies in Big Data”中。该系列旨在快速发布大数据领域的新进展和重要成果，重点关注大数据理论、研究、开发和应用，这些都与工程、计算机科学、物理学、经济学以及生命科学等多学科紧密相连。书中关注的对象是源自各种数字化源的大规模、复杂和/或分布式的数据集，这些数据来源于传感器、物理仪器的测量结果，模拟实验、众包项目、社交网络活动，以及互联网交易记录，如电子邮件和视频点击流等。作者们深入探讨了如何分析和理解这些海量数据，以支持计算智能的研究，包括神经网络、进化计算、软计算等领域。该卷作为系列的一部分，不仅包含原创的研究论文，也涵盖了教学讲义和编纂的合集，全面展示了大数据处理在当今科技环境中的关键作用。读者可以从中学到如何利用大数据技术改进决策过程、优化业务流程、挖掘潜在价值，以及解决复杂问题的方法。每一章节都反映了当前大数据研究的前沿趋势和技术实践，对于那些对大数据感兴趣的专业人士、研究人员和学生来说，是一本极具价值的学习资料和参考资料。通过阅读Studies in Big Data 03，读者不仅可以提升对大数据技术的理解，还能了解到如何将其应用于实际场景，推动科技进步和社会发展。系列网站 <http://www.springer.com/series/11970> 提供了更多该系列的后续出版物信息，持续追踪并分享大数据领域的最新研究成果。

8 K. Yada

mining. When trying to extract useful characteristics from a huge volume of text

data, dimensionality reduction, which narrows characteristics down to important key

words, is one of the most important technologies. This chapter describes a dimen-

sionality reduction technique which uses rare terms and co-occurring terms. Chapter,

“Panel Data Analysis Via Variable Selection and Subject Clustering” studies cluster-

ing techniques for panel data used in marketing and economics. This chapter proposes

a technique which does variable selection and clustering using BIC at the same time,

and describes results when this is applied to U.S. company survey data.

Part II, “Knowledge Discovery from Text”, describes research on knowledge dis-

covery from huge text data. One can create various new services from huge amounts

of text information, such as Internet comments, call center conversation recordings,

or an academic database. This part describes cases in which text mining was applied

to services, and explains the core technologies used.

Chapters, “A Weighted Density-Based Approach for Identifying Standardized

Items that are Signiﬁcantly Related to the Biological Literature”, “Nonnegative

Tensor Factorization of Biomedical Literature for Analysis of Genomic Data” and

“Text Mining of Business-Oriented Conversations at a Call Center”, also in Part

II, discuss research on knowledge discovery from biological literature. Chapter,

“A Weighted Density-Based Approach for Identifying Standardized Items that are

Signiﬁcantly Related to the Biological Literature” studies techniques using abstracts

of published biological literature to classify their content. The authors propose

a weighted density-based approach, and found that this can be used to classify

papers more accurately than previous techniques. Chapter, “Nonnegative Tensor

Factorization of Biomedical Literature for Analysis of Genomic Data” discusses a

more speciﬁc issue, by describing a technique using Non-negative Tensor Factoriza-

tion (NTF) to extract relationships between genes and transcription factors. Chapter,

“Text Mining of Business-Oriented Conversations at a Call Center” describes a case

in which text mining was applied to data on conversations between operators and

customers in a call center. The authors present a framework using a Support Vector

Machine (SVM), and provide a technique which extracts conversational expressions

and words which affect the customer’s situation.

Part III, “Approach for New Services in Social Media”, contains research that

apply data mining to data accumulated in Internet social networking services which

have grown rapidly in recent years, to create new knowledge and services. Chapter,

“Scam Detection in Twitter” discusses scam detection from Twitter. This chapter

proposes a scam detection technique using sufﬁx trees which combine Twitter’s

unique data content, where one tweet is a short sentence using a special symbol such

as “#”. Chapter, “A Matrix Factorization Framework for Jointly Analyzing Multiple

Nonnegative Data Sources” describes a search method using combined data from

multiple social media, applies this to data from YouTube and BlogSpot, and shows its

usefulness. Social networking services providing various services exist, but it is also

possible to combine these to create other new services. Chapter, “Recommendation

Systems for Web 2.0 Marketing” proposes a framework which combines multiple

data sources, and applies them to a recommendation system. It is difﬁcult for the

information needed for collaborative ﬁltering to be gathered only in a company’s

Data Mining for Service 9

own Internet shop search data. This chapter proposes a framework which combines

product search information and social media information.

Part IV, “Data Mining Spreading into Various Service Fields”, describes lead-

ing research ﬁelds in which data mining is applied to diverse data obtained from

new information devices, with various services being created. Chapter, “Handling

Imbalanced and Overlapping Classes in Smart Environments Prompting Dataset”

describes a case where data mining is applied to smart home environments based on

home electronic products connected to networks, a case which has spread rapidly in

recent years. Data created in actual events often contains imbalanced data, such as

mistaken use of home electronics. This chapter proposes ClusBUS, a clustering tech-

nique to handle the overlapping class problem created by imbalanced data. Chapter,

“Change Detection from Heterogeneous Data Sources” studies a technique of change

detection, focused on sensor data which has been attracting the most attention as big

data in recent years. The authors describe the singular spectrum transformation tech-

nique for change-point detection, which handles dynamic characteristics in actual

heterogeneous sensor data, and shows its usefulness. Chapter, “Interesting Subset

Discovery and Its Application on Service Processes” describes applied cases of

data mining using various databases accumulated in companies. It shows speciﬁc

cases where data mining is applied to databases which mix continuous values and

discrete values: employee satisfaction surveys, IT support system failure surveys,

data processing performance surveys for data centers, etc. Chapter, “Text Document

Cluster Analysis Through Visualization of 3D Projections” concludes with prob-

lems of visualization, which most affect the performance of data mining in business

applications. This chapter shows a framework and system to visualize the clustering

process and results in text mining, so non-technical users can also understand the

phenomena.

References

1. Chesbrough, H., Spohrer, J.: A research manifesto for service science. Commun. ACM 49(7),

35–40 (2006)

2. Fisk, R.P., Grove, S.J., John, J.: Interactive Services Marketing, 3rd edn. Houghton Mifﬂin

Company, Boston (2007)

3. Hamuro, Y., Katoh, N., Yada, K.: MUSASHI: ﬂexible and efﬁcient data preprocessing tool for

KDD based on XML. In: DCAP2002 Workshop Held in Conjunction with ICDM2002, pp.

38–49 (2002)

4. Hui, S.K., Fader, P.S., Bradlow, E.T.: Path data in marketing: an integrative framework and

prospectus for model building. Mark. Sci. 28(2), 320–335 (2009)

5. Ichikawa, K., Yada, K., Washio, T.: Development of data mining platform MUSASHI towards

service computing. In: 2010 IEEE International Conference on Granular Computing (GrC),

pp. 235–240 (2010)

6. Ohsawa, Y., Yada, K. (eds.): Data Mining for Design and Marketing. CRC Press, USA (2009)

7. Sasser, W.E., Olsen, R.P., Wyckoff, D.D.: Management of Service Operations, Allyn and Bacon,

Boston (1978)

8. Spohrer, J., Maglio, P.P., Bailey, J., Gruhl, D.: Steps toward a science of service systems. IEEE

Comput. 40(1), 71–77 (2007)

剩余290页未读，继续阅读

liucg9999

粉丝: 2
资源: 15

大数据挖掘服务研究进展——Studies in Big Data Vol.3

Studies in Big Data 02

Studies in Big Data 01

Python for Programmers: with Big Data and Artificial Intelligence Case Studies

Embracing social big data in wireless system design

Big Data in Omics and Imaging: Association Analysis

英文原版-The Intelligent Enterprise in the Era of Big Data 1st Edition

Big Data, Data Mining, and Machine Learning

Big Data and Computational Intelligence in Networking-CRC(2018).pdf

Big Data Management and Processing

big data analytics with java

最新资源