Customer analytics

The proliferation of distributed cloud environments necessitates innovative approaches to seamless automation, particularly through the integration of intelligent systems and emerging integration technologies. This study explores an architectural framework that leverages artificial intelligence (AI), container orchestration, and data-driven decision-making to optimize workflows across geographically dispersed cloud infrastructures. This research highlights the transformative potential of multi-cloud strategies, federated learning, and service mesh paradigms in realizing this vision. By incorporating quantitative metrics and real-world case studies, we provide empirical evidence on reduced latency, improved system resilience, and enhanced resource utilization, fostering a paradigm shift in intelligent cloud automation.

I wish to also express my company supervisor André Rufino for the opportunity to undergo this internship as well as all his dedication, teachings and patience in ensuring that this project was carried out as smoothly as possible. I would also like to thank Anwar El-Jawhari for his guidance, dedication, for supervising me throughout the project and passing on his incredible data science knowledge. My gratitude also goes towards Diana Duarte, who was my assigned business specialist, for ensuring that my integration into the company and into the insurance world was easily achieved. I would also like to thank my Master's colleagues Andrew Talas and Jorge Abreu who were with me throughout the internship journey, for their help, friendship and support. Furthermore, I would like to thank everyone at the host company and a special thank you to my team who contributed to making this learning experience as pleasant as possible.

The paper outlines the outcomes of an empirical investigation of the links between involvement of insurance firms in customer analysis and their financial performance. The data were collected through a CAWI survey of independent insurance agents in Poland (n=590). Statistical methods included exploratory factor analysis and structural equation modeling. In keeping with the customer lifetime value model and loyalty management theory the concept of customer analysis was operationalized as a second-order reflective construct, expressed through five first order subconstructs. Financial performance was measured with an index of five self-reported metrics. The findings lend support to the hypothesis of a positive association between engaging in customer analysis and financial performance.

With the triggering effect of Covid-19 pandemic, the role of digitalization has become a strategic target and expedited the digital transformation process. World's direction to the digital future has therefore shaped the use of newage technologies, such as internet of things, artificial intelligence (AI), machine learning and blockchain. In response to this evolvement of new-age technologies, a noticeable shift from data-driven analysis to technology-oriented applications has occurred, particularly addressing the significance of analytics and AI. These rapid advancements of AI applications influence the use of customer analytics whilst enhancing the importance both for the general understanding and individual behavior of customers, within the scope of customer analytics. Considering the embeddedness of these technologies on practical applications, this study acknowledges the high-impact role and power of AI. In this regard, the study concentrates AI applications from the perspectives of customer analytics. Furthermore, the task of AI, the level of intelligence of AI applications and how the information from customer analytics is obtained and exploited by these applications are discussed. Focusing on the practical case applications, the study suggests a taxonomical structure of AI and customer analytics.

Purpose – Mobile checkout in the retail store has the promise to be a rich source of big data. It is also a means to increase the rate at which big data flows into an organization as well as the potential to integrate product recommendations and promotions in real time. However, despite efforts by retailers to implement this retail innovation, adoption by customers has been slow. The paper aims to discuss these issues. Design/methodology/approach – Based on interviews and focus groups with leading retailers, technology providers, and service providers, the authors identified several emerging in-store mobile scenarios; and based on customer focus groups, the authors identified potential drivers and inhibitors of use. Findings – A first departure from the traditional customer checkout process flow is that a mobile checkout involves two processes: scanning and payment, and that checkout scenarios with respect to each of these processes varied across two dimensions: first, location – wh...

The paper outlines the outcomes of an empirical investigation of the links between involvement of insurance firms in customer analysis and their financial performance. The data were collected through a CAWI survey of independent insurance agents in Poland (n=590). Statistical methods included exploratory factor analysis and structural equation modeling. In keeping with the customer lifetime value model and loyalty management theory the concept of customer analysis was operationalized as a second-order reflective construct, expressed through five first order subconstructs. Financial performance was measured with an index of five self-reported metrics. The findings lend support to the hypothesis of a positive association between engaging in customer analysis and financial performance.

The paper outlines the outcomes of an empirical investigation of the links between involvement of insurance firms in customer analysis and their financial performance. The data were collected through a CAWI survey of independent insurance agents in Poland (n=590). Statistical methods included exploratory factor analysis and structural equation modeling. In keeping with the customer lifetime value model and loyalty management theory the concept of customer analysis was operationalized as a second-order reflective construct, expressed through five first order subconstructs. Financial performance was measured with an index of five self-reported metrics. The findings lent support to the hypothesis of a positive association between engaging in customer analysis and financial performance.

We argue in this paper that in order to properly capture opinion and sentiment expressed in texts or dialogs any system needs a deep linguistic processing approach. As in other systems, we used ontology matching and concept search, based on standard lexical resources, but a natural language understanding system is still required to spot fundamental and pervasive linguistic phenomena. We implemented these additions to VENSES system and the results of the evaluation are compared to those reported in the state-of-the-art systems in sentiment analysis and opinion mining. We also provide a critical review of the current benchmark datasets as we realized that very often sentiment and opinion is not properly modeled.

T ree-splitting algorithms were initially developed by John Sonquist and James Morgan (1964) and rendered into THAID by Morgan and R.C. Messenger (1973) of the Institute for Social Research at the University of Michigan. G.V. Kass transformed this into CHAID (Chi-square Automatic Interaction Detection) in 1980. Leo Breiman and Jerome Friedman independently worked on the innovation that became CART (Classification and Regression Trees) in the early 1980s. Since then, those algorithms and others have been developed into viable tree-splitting computational applications by SPSS, Statistical Innovations (SI), Salford Systems, SAS, Insightful Corp., and others. The introduction of SI-CHAID 4.0 is much more than just another entrant to this field; it is a creative approach to handling the problem of effectively segmenting databases when there are multiple dependent variables, possibly of mixed types, and a large number of candidate predictor variables. In addition, SI-CHAID 4.0 (SIC4), alo...

English. We present results obtained by our system ITGetaruns for all tasks. It is a linguistic rule-based system in its bottom-up version that computes a complete parser of the input text. On top of that it produces semantics at different levels which is then used by the algorithm for sentiment and polarity detection. Our results are not remarkable apart from the ones related to Irony detection, where we ranked fourth over eight participants. The results were characterized by our intention to favour Recall over Precision and this is also testified by Recall values for Polarity which in one case rank highest of all. Italiano. Presentiamo i risultati ottenuti dal nostro sistema ITGetaruns per tutti i task. Si tratta di un sistema basato su regole linguistiche nella sua versione bottom-up, che produce un parse complete del testo in ingresso. Al di sopra di questo produce semantica a diversi livelli, che viene poi usata dall'algoritmo per l'analisi della polarita e della sogget...

Purpose
Mobile checkout in the retail store has the promise to be a rich source of big data. It is also a means to increase the rate at which big data flows into an organization as well as the potential to integrate product recommendations and promotions in real time. However, despite efforts by retailers to implement this retail innovation, adoption by customers has been slow. The paper aims to discuss these issues.

Design/methodology/approach
Based on interviews and focus groups with leading retailers, technology providers, and service providers, the authors identified several emerging in-store mobile scenarios; and based on customer focus groups, the authors identified potential drivers and inhibitors of use.

Findings
A first departure from the traditional customer checkout process flow is that a mobile checkout involves two processes: scanning and payment, and that checkout scenarios with respect to each of these processes varied across two dimensions: first, location – whether they were fixed by location or mobile; and second, autonomy – whether they were assisted by store employees or unassisted. The authors found no evidence that individuals found mobile scanning to be either enjoyable or to have utilitarian benefit. The authors also did not find greater privacy concerns with mobile payments scenarios. The authors did, however, in the post hoc analysis find that mobile unassisted scanning was preferred to mobile assisted scanning. The authors also found that mobile unassisted scanning with fixed unassisted checkout was a preferred service mode, while there was evidence that mobile assisted scanning with mobile assisted payment was the least preferred checkout mode. Finally, the authors found that individual differences including computer self-efficacy, personal innovativeness, and technology anxiety were strong predictors of adoption of mobile scanning and payment scenarios.

Originality/value
The work helps the authors understand the emerging mobile checkout scenarios in the retail environment and customer reactions to these scenarios.

We argue in this paper that in order to properly capture opinion and sentiment expressed in texts or dialogs any system needs a deep linguistic processing approach. As in other systems, we used ontology matching and concept search, based on standard lexical resources, but a natural language understanding system is still required to spot fundamental and pervasive linguistic phenomena. We implemented these additions to VENSES system and the results of the evaluation are compared to those reported in the state-of-the-art systems in sentiment analysis and opinion mining. We also provide a critical review of the current benchmark datasets as we realized that very often sentiment and opinion is not properly modeled.

In this paper, we present our solution for pragmatic analysis of call center conversations in order to provide useful insights for enhancing Call Center Analytics to a level that will enable new metrics and key performance indicators (KPIs) beyond the standard approach. These metrics rely on understanding the dynamics of conversations by highlighting the way participants discuss about topics. By doing that we can detect situations that are simply impossible to detect with standard approaches such as (un)cooperative behaviors, controversial topics, customer oriented behaviors and also predict customer satisfaction.

We define a new task of question answering on meeting records and assess its difficulty in terms of types of information and retrieval techniques required. The importance of this task is revealed by the increasingly growing interest in the design of sophisticated interfaces for accessing meeting records such as meeting browsers. We ground our work on the empirical analysis of elicited user queries. We assess what is the type of information sought by the users and perform a user query classification along several semantic dimensions of the meeting content. We found that queries about argumentative processes and outcomes represent the majority among the elicited queries (about 60%). We also assess the difficulty in answering the queries and focus on the requirements of a prospective QA system to successfully deal with them. Our results suggest that standard Information Retrieval and Question Answering alone can only account for less than 20% of the queries and need to be complemented with additional type of information and inference.

Retailing has drastically transformed business and customer now have access to wide range of products offered through retail outlets both in organized and unorganized sector. In order to remain highly competitive and sustain growth, retailing companies are formulating marketing strategies based on data. This has navigated to a paradigm shift in retailing, where data is seen as an asset to the firm in understanding timely needs of customers, predicting buying behavior, and proposing performance metrics to assess effectiveness. Retailing companies are finding ways to extract meaningful information from larger datasets that are generated through different sources, in different formats. As we see Big Data is one of the buzzword taking technology world by storm, retailing companies are trying to understand how the use of Big Data Analytics can empower them to take right decisions. This paper investigates how the use of Big Data analytics impacts buying trends in retail industry.

Telecommunication companies are functioning under highly competitive and challenging environment. It collects large amount of data from various operational systems like call detail data, network data, customer data etc. which is used to solve many problems. It required mining to improve their efficiency such as marketing, fraud identification and manage their networks. Due to enormous size of data sets, sequential temporal aspects of their data these industries face number of challenges. However, this knowledge is not given explicitly; it is buried in huge amount of data. Advanced methods and algorithms are being developed for knowledge extracting. In this paper we will discuss on using data mining for solving selected problems in telecommunication industry.

In this paper we present two new mechanisms we created in VENSES, the system for semantic evaluation of the University of Venice. The first mechanism is used to match predicate-argument structures with different governors, a verb and a noun, respectively in the Hypothesis and the Text. It can be defined Augmented Finite State Automata (FSA) which are matching procedures based on tagged words in one case, and dependency relations in another. In both cases, a number of inferences -the augmentation -is fired to match different words. The second mechanism is based on the output of our module for anaphora resolution. Our system produces antecedents for pronominal expressions and equal nominal expressions. On the contrary, no decision is taken for "bridging" expressions. So the "bridging" mechanism is activated by the Semantic Evaluator and has access to the History List and the semantic features associated to each referring expression. If constraint conditions meet, the system looks for a similar association of property/entity in web ontologies like Umbel, Yago and DBPedia. The two mechanisms have been proven to contribute a 5% and 3% accuracy, respectively.

Resumen: The system for semantic evaluation VENSES (Venice Semantic Evaluation System) is organized as a pipeline of two subsystems: the first is a reduced version of GETARUN, our system for Text Understanding. The output of the system is a flat list of head-dependent structures (HDS) with Grammatical Relations (GRs) and Semantic Roles (SRs) labels. The evaluation system is made up of two main modules: the first is a sequence of linguistic rule-based subcalls; the second is a quantitatively based measurement of input ...

The paper explores the relationship between conducting customer analysis and financial performance. The data for the study were collected from 590 small insurance intermediaries in the Polish market. The structural equation modelling indicates that the strongest predictor of financial performance was the use of formalized knowledge processing, followed by the scope of performed customer analysis. Other factors positively correlated with financial standing included: earning most revenues from corporate clients (versus consumers) and employing policies aimed at regaining former customers. The study also finds that add-on on selling was not significantly associated with better financial results.

The paper outlines the outcomes of an empirical investigation of the links between involvement of insurance firms in customer analysis and their financial performance. The data were collected through a CAWI survey of independent insurance agents in Poland (n=590). Statistical methods included exploratory factor analysis and structural equation modeling. In keeping with the customer lifetime value model and loyalty management theory the concept of customer analysis was operationalized as a second-order reflective construct, expressed through five first order subconstructs. Financial performance was measured with an index of five self-reported metrics. The findings lend support to the hypothesis of a positive association between engaging in customer analysis and financial performance.

We argue in this paper that in order to properly capture opinion and sentiment expressed in texts or dialogs any system needs a deep linguistic processing approach. As in other systems, we used ontology matching and concept search, based on standard lexical resources, but a natural language understanding system is still required to spot fundamental and pervasive linguistic phenomena. We implemented these additions to VENSES system and the results of the evaluation are compared to those reported in the state-of-the-art systems in sentiment analysis and opinion mining. We also provide a critical review of the current benchmark datasets as we realized that very often sentiment and opinion is not properly modeled.

In this paper we present two new mechanisms we created in VENSES, the system for semantic evaluation of the University of Venice. The first mechanism is used to match predicate-argument structures with different governors, a verb and a noun, respectively in the Hypothesis and the Text. It can be defined Augmented Finite State Automata (FSA) which are matching procedures based on tagged words in one case, and dependency relations in another. In both cases, a number of inferences -the augmentation -is fired to match different words. The second mechanism is based on the output of our module for anaphora resolution. Our system produces antecedents for pronominal expressions and equal nominal expressions. On the contrary, no decision is taken for "bridging" expressions. So the "bridging" mechanism is activated by the Semantic Evaluator and has access to the History List and the semantic features associated to each referring expression. If constraint conditions meet, the system looks for a similar association of property/entity in web ontologies like Umbel, Yago and DBPedia. The two mechanisms have been proven to contribute a 5% and 3% accuracy, respectively.

Eine Google Suche nach Customer Analytics bringt mehr als 60 Millionen Ergebnisse zurück. Das Thema ist nicht unbekannt und sehr präsent in der Fachliteratur. Alle "Big Player" der IT Branche (Software und Beratung) haben neue Angebote / Produkte in diesem Bereich und haben das Thema auf der Tagesordnung. Aber eigentliche Kundenanalysen sind nichts neues -was ist Heute anders als in der Vergangenheit? Der Vortrag wird die heutigen Möglichkeiten der Analyse darstellen, die bis vor weniger Jahren undenkbar waren (weil nur für Grosskonzerne verfügbar,oder weil es neue Datenquellen wie Social Media noch nicht gab) und die Integration der Ergebnisse (z.T. in Echtzeit) in den Geschäftsprozessen. Dadurch ist eine effektivere Steuerung der Front-Office Prozesse und teilweise der Produktentwicklung möglich. Fallstudien zeigen Anwendungen in verschiedenen Branchen, um mögliche Implementierungsbeispiele konkret darzustellen.

We argue in this paper that in order to properly capture opinion and sentiment expressed in texts or dialogs any system needs a deep linguistic processing approach. As in other systems, we used ontology matching and concept search, based on standard lexical resources, but a natural language understanding system is still required to spot fundamental and pervasive linguistic phenomena. We implemented these additions to VENSES system and the results of the evaluation are compared to those reported in the state-of-the-art systems in sentiment analysis and opinion mining. We also provide a critical review of the current benchmark datasets as we realized that very often sentiment and opinion is not properly modeled.

In this chapter we define the notion of multimedia conversational system and provide a classification schema for characterizing the type of content which can be produced, stored and retrieved with these systems. This classification schema will be used in reviewing the capabilities of a number of existing multimedia conversational systems and to assess requirements for advanced indexing and retrieval of conversational content. To meet these requirements, we present new types of indexing techniques and provide an evaluation of their effectiveness on real case studies.

... Raúl Ruggia (Universidad de la República, Uruguay), Flavia Serra (Universidad de la República, Uruguay), and Elena Martirena (Universidad ... Coming Soon! ... OpAL: A System for Mining Opinion from Text for Business Applications (pages 147-177). Alexandra Balahur (University ...

We define a new task of question answering on meeting records and assess its difficulty in terms of types of information and retrieval techniques required. The importance of this task is revealed by the increasingly growing interest in the design of sophisticated interfaces for accessing meeting records such as meeting browsers. We ground our work on the empirical analysis of elicited user queries. We assess what is the type of information sought by the users and perform a user query classification along several semantic dimensions of the meeting content. We found that queries about argumentative processes and outcomes represent the majority among the elicited queries (about 60%). We also assess the difficulty in answering the queries and focus on the requirements of a prospective QA system to successfully deal with them. Our results suggest that standard Information Retrieval and Question Answering alone can only account for less than 20% of the queries and need to be complemented with additional type of information and inference.

The volume of information circulating in a typical enterprise continues to increase. Knowledge hidden in the information however, is not fully utilized, as most of the information is described in textual form (as sentences). A large amount of text information can be analyzed objectively and efficiently with Text Mining.The field of text mining has received a lot of attention due to the ever increasing need for managing the information that resides in the vast amount of available text documents. Text documents are characterized by their unstructured nature. Ever increasing sources of such unstructured information include the World Wide Web, biological databases, news articles, emails etc.

In this chapter we present the major challenges of a new trend in business analytics, namely Interaction Mining. With the proliferation of unstructured data as the result of people interacting with each other using digital networked devices, classical methods in text business analytics are no longer effective. We identified the causes of their failure as being related to the inadequacy of dealing with conversational data. We propose then to move from Text Mining towards Interaction Mining, and we make several cases for this transition in areas such as marketing research, social media analytics, and customer relationship management. We also propose a roadmap for the future development of Interaction Mining by challenging the current practices in business intelligence and information visualization.

We argue in this paper that in order to properly capture opinion and sentiment expressed in texts or dialogs any system needs a deep linguistic processing approach. As in other systems, we used ontology matching and concept search, based on standard lexical resources, but a natural language understanding system is still required to spot fundamental and pervasive linguistic phenomena. We implemented these additions to VENSES system and the results of the evaluation are compared to those reported in the state-of-the-art systems in sentiment analysis and opinion mining. We also provide a critical review of the current benchmark datasets as we realized that very often sentiment and opinion is not properly modeled.