Customer Technical Support Research Articles

Preprocessing Text Data for the Doc2Vec Model in the Automatic Classification of Tickets in Support Centers of Software Providers

Leading software providers typically implement customer technical support functions, which are crucial for promoting and enhancing the competitiveness of their products and services in global markets. The high volume and heterogeneity of support tickets (functional, temporal, linguistic, etc.) highlight the importance of efficient classification systems. Effective classification optimizes the distribution of these tickets among support center specialists and automates their processing using an established knowledge base. However, classifying these tickets is a loosely formalized task. For companies that have accumulated substantial data on customer requests, automating classification through machine learning methods and natural language processing models, such as Word2Vec, FastText, BERT, and GPT, becomes feasible. It is generally accepted that classification effectiveness primarily depends on the model employed. Nevertheless, the quality of these models is significantly influenced by the nature of the training data. Literature review of the reveals significant research interest in methods for the automatic classification of tickets specifically tailored to the operational conditions of software provider support centers. However, there is a noticeable gap in the literature regarding the impact of data preprocessing on the quality of these models. The article aims to clarify the techniques of data preprocessing and analyze their impact on the effectiveness of text classification, considering the specificity of software provider support centers. This study examines the stages of the automatic classification process for tickets, accounting for the unique characteristics of the data (customer text requests). A relevant set of specified methodological and instrumental tools was developed and tested using open data from a global software provider (DevExpress). The testing involved a database of 165,000 tickets. The study's results indicate that preprocessing can improve classification metrics such as F-measure, Precision, and Recall from 77% to 79%. Additionally, preprocessing significantly reduces the dimensionality of text data (by 48.2%) and increases model training speed (by 26.5%) without loss of accuracy, ensuring cost-efficiency and operational efficiency in the use of computational resources.

Construction and Risk Analysis of Marketing System Based on Deep Neural Network

Enterprises needs to realize the unity of economic and social benefits. From the marketing theory, we know that the economic and social benefits form a symbiotic effect, which determines that enterprises need to pay attention to their own service quality. The current risks are in the areas of “customer relationship management” and “customer technical support,” specifically, the inability to effectively meet the needs of customers in terms of corporate infrastructure and product details, as well as the increase in costs for other customer services. The article firstly completes the construction of marketing risk early warning system, analyzes marketing risk types and formation factors, constructs indicators and mathematical modeling, then proposes the ANN combined with Petri algorithm, which is used to realize marketing risk early warning, completes sample selection and indicator quantification, sets up the early warning model, and conducts false alarm and false alarm degree analysis.

Respond, restore, resolve: Achieving 7-nines availability telecommunications systems in the field

Achieving telecommunications systems with 7-nines (0.9999999) availability requires taking a 5-nines telecommunications system and providing extremely responsive and technically excellent customer support. 5-nines telecommunications systems have robust hardware and software reliability assets and are designed to meet demanding serviceability and maintainability requirements. These assets are characterized by extremely low failure rates, fast recovery times, and a high automatic recovery coverage rate. However, to surpass 5-nines availability, excellent human observation and intervention are required. The theme “respond, restore, resolve” is the jingle of an important customer support reliability pattern. It is meant to help remind field support engineers of the best-in-class overall strategy for responding to customer issues. This and other important customer support patterns that impact reliability will be described in detail. The audience for the paper includes members of customer technical support organizations as well as telecommunication system architects and designers who hope to build systems capable of achieving very high availability.

Using case-based retrieval for customer technical support

Cascade, a case-based interactive system that helps technical support engineers solve the problems of customers whose hardware or software has stopped working, is discussed. Cascade has a case library of failures of VMS device drivers and suggests solutions to new failures. The system uses validated retrieval to address aspects of failure recovery that are specific to help desks. Cascade's implementation, which was accomplished by developing a seed case base, a knowledge base containing knowledge about similarities between surface-feature values, and a validation model containing the necessary probes and knowledge about them, is described. >