Textual data mining of service center call records

Tan, Pang-Ning; Blau, Hannah; Harp, Steven A.; Goldman, Robert P.

doi:10.1145/347090.347177

Cited by 30 publications

(15 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Indeed, Tan et al [27] have extracted useful feature-level information from service center call records-that contain both fixed-format (i.e., quantitative) and free-text (i.e., qualitative) data-and, through clustering techniques, determined the cost of different types of service requests. They also found that by incorporating qualitative information, a better classification system resulted, thus underscoring the importance of fusing nonhomogeneous data, especially from the perspective of enhancing the production and delivery of emerging services.…”

Section: Research Areamentioning

confidence: 99%

“…Interestingly, however, the coincident effort may actually yield a synergistic dividend. It is possible that the fused nonhomogeneous data could enhance the real-time analysis of that data, especially in regard to the issue of whether the new datum is an outlier or part of a new phenomenon; that is, the extra dimensionality of the fused data could very well help resolve the issue, much like how the fused data helped Tan et al [27] with their classification scheme. However, even with such a dividend, considerable research remains.…”

Section: Research Areamentioning

confidence: 99%

See 1 more Smart Citation

Toward a decision informatics paradigm: a real-time, information-based approach to decision making

Tien

2003

IEEE Trans. Syst., Man, Cybern. C

View full text Add to dashboard Cite

The focus of this paper is on decision making; more specifically, on what decision making requirements are needed in the future. We augur for a decision informatics paradigm; it is a real-time, information-based approach to decision making. The paradigm is supported by two sets of technologies (i.e., information and decision technologies) and underpinned by three disciplines (i.e., data fusion/analysis, decision modeling, and systems engineering). We begin by considering the context-and needs-for decision making as the economies of the world change and evolve, especially in regard to emerging services; then our proposed decision informatics paradigm is detailed and illustrated, together with an in-depth review of a critical, underpinning research area (dealing with real-time fusion and analysis of multiple nonhomogeneous data sources), followed by several concluding remarks.Index Terms-Data analysis, data fusion, decision making, mass customization, quantitative modeling, systems engineering. I. CONTEXTUAL CONSIDERATIONST HE focus of this paper is on decision making; more specifically, on what decision making requirements are needed in the future. We augur for a decision informatics paradigm; it is a real-time, information-based approach to decision making. We begin by considering the context-and needs-for decision making as the economies of the world change and evolve (Section I), especially in regard to emerging services (Section II); then our proposed decision informatics paradigm is detailed and illustrated (Section III), together with an in-depth review of a critical, underpinning research area (Section IV), followed by several concluding remarks (Section V).In order to provide a context for considering current and future decision making requirements, it is instructive to review the critical stages in a nation's economic evolution [32]. As summarized in Table I, there have been thus far three stages. The first-mechanical-stage focuses on agriculture and mining (i.e., living off of the land, air, and sea); it seeks to enhance farming productivity, employs mechanical tools that have product life cycles on the order of decades, depends mostly on muscle power, embraces a living standard concerned with subsistence, and is limited in its scope of economic influence-primarily impacting the family or immediate locale. The second-electrical-stage focuses on manufacturing and construction (i.e., creating and producing products and structures); it seeks to enhance factory productivity, employs electromechanical machines that have product life cycles on the order of years, depends on both muscle and brain power, embraces a living standard concerned with the quality of goods, and is broader in its scope of economic influence-impacting the region or nation. The third-information-stage focuses on services (i.e., creating and delivering added value that is essentially intangible); it seeks to enhance information (i.e., processed data from a decision making perspective) productivity, employs information (including communicat...

show abstract

Section: Research Areamentioning

confidence: 99%

Section: Research Areamentioning

confidence: 99%

Toward a decision informatics paradigm: a real-time, information-based approach to decision making

Tien

2003

IEEE Trans. Syst., Man, Cybern. C

View full text Add to dashboard Cite

show abstract

“…Unlike most of the previous works that dealt primarily with quantitative data, the study by Tan et al 23 investigated service center call records, comprising both textual and fixed-format columns, to extract information about the expected cost of different kinds of service requests. They found that the incorporation of information from free-text fields provided a better categorization of these records, thus facilitating better predictions of the cost of the service calls.…”

Section: Previous Work On Data Mining Within the Pdpmentioning

confidence: 99%

“…Although this service center database is quite similar in structure to that analysed by Tan et al 23 , the datamining operation used is different. In the present study, association analysis is used, as opposed to classification analysis used in the earlier study.…”

Section: Service Center Database-case Studymentioning

confidence: 99%

The Needs and Benefits of Applying Textual Data Mining within the Product Development Process

Menon

Tong

Sathiyakeerthi

et al. 2003

Quality & Reliability Eng

View full text Add to dashboard Cite

As a result of the growing competition in recent years, new trends such as increased product complexity, changing customer requirements and shortening development time have emerged within the product development process (PDP). These trends have added more challenges to the already-difficult task of quality and reliability prediction and improvement. They have given rise to an increase in the number of unexpected events in the PDP. Traditional tools are only partially adequate to cover these unexpected events. As such, new tools are being sought to complement traditional ones. This paper investigates the use of one such tool, textual data mining for the purpose of quality and reliability improvement. The motivation for this paper stems from the need to handle 'loosely structured textual data' within the product development process. Thus far, most of the studies on data mining within the PDP have focused on numerical databases. In this paper, the need for the study of textual databases is established. Possible areas within a generic PDP for consumer and professional products, where textual data mining could be employed are highlighted. In addition, successful implementations of textual data mining within two large multinational companies are presented. Copyright c 2003 John Wiley & Sons, Ltd.KEY WORDS: product development process; textual databases; service center; call center; data mining PRODUCT DEVELOPMENT PROCESSA product development process (PDP) is the sequence of steps or activities which an enterprise employs to conceive, design and commercialize a product 1 . A well-organized and coherent development process serves to ensure the efficient delivery of a final product that suits a customer's wants. Such products are truly the lifeblood of a company's long term economic existence. As such, it is no surprise that companies are

show abstract

“…In one study 2 of contact records at Honeywell, the researchers looked at ways to classify both structured and unstructured data. The structured data included the various times the customer spoke with a technical support engineer at Honeywell, the customer name and other fixed data like call time, while the free text entered by the Honeywell engineer was unstructured.…”

Section: Differentiating Between Structured and Unstructured Data--homentioning

confidence: 99%

Data Mining in Call Centers: The Overlooked Interaction between Employees

Lohse¹,

Sanati-Mehrizy

Minaie

2015 ASEE Annual Conference and Exposition Proceedings

View full text Add to dashboard Cite

Many data mining techniques have been applied to technical support call center interactions between customers and employees. There remains a vast, largely untapped source of data related to intraemployee interactions. In typical call centers, there are chat rooms for more experienced first-level support technicians to help newer, less experienced technicians. In addition, when a first level technician needs more assistance, they will contact a second-level employee by phone or chat.Mining these records of inter-employee interaction could lead to improved interaction and eventually to a machine learning system that can substitute for the second-level employee in the initial stage of assisting the less trained first-level employees. One of the authors has worked in several technical support call centers and therefore has detailed knowledge of their operations and areas that can be improved. This paper was inspired by this author noticing much of the communication between various levels of employees is not retained or analyzed, leading to a phenomenon of repeating and recreating the same efforts day after day, training class after training class. This paper proposes a system of studying the interactions between first-and second-level technical support employees using data mining and machine learning techniques. The proposed system would be inserted into the normal flow of information between first-level tech support employees and the second-level co-workers they consult with when they need help. The overall goal of this proposal is to intercept solutions that are provided on an ad-hoc basis and capture them into a corporate database that will present these solutions in the future, as they are learned. This type of system would be flexible enough to adapt to new products and new questions as new questions are posed to second-level experts.

show abstract

Textual data mining of service center call records

Cited by 30 publications

References 10 publications

Toward a decision informatics paradigm: a real-time, information-based approach to decision making

Toward a decision informatics paradigm: a real-time, information-based approach to decision making

The Needs and Benefits of Applying Textual Data Mining within the Product Development Process

Data Mining in Call Centers: The Overlooked Interaction between Employees

Contact Info

Product

Resources

About