Large language models (LLMs) have revolutionized natural language processing, enabling unprecedented capabilities in reasoning, planning, and tool utilization. Enterprises are increasingly adopting LLM-powered agents to automate complex workflows, from meeting summarization (e.g., Microsoft Copilot) to supply chain optimization and customer service orchestration. However, deploying agentic AI systems in enterprise settings introduces unique challenges, including decision making under uncertainty, multi-agent collaboration, security vulnerabilities, and trust gaps in mission-critical applications. This workshop aims to bridge the gap between academia and industry to explore LLM-driven agentic systems tailored for enterprise needs. We focus on three pillars: 1) emerging architectures that enable dynamic task decomposition and tool invocation; 2) domain-specific applications such as case studies in supply chain and employee productivity domain; 3) evaluation and governance such as the AAEF (Agentic Application Evaluation Framework) and security strategies.

Nowadays, while most machine learning research on customer journey optimization has focused on short-term success metrics such as click-through rates and optimal ad placement, there has been little consideration given to developing a coherent system for end-to-end customer journey optimization. Such a system would encompass all aspects of the customer experience, from presenting the right product value to the right users, to understanding a user's likelihood of conversion and long-term value to the platform, as well as their propensity for cross-selling and risk of churning. Currently, models and algorithms for customer journey optimization are often developed in isolation, leading to inefficiencies in modeling and data pipelines. Furthermore, the customer is often viewed as a collection of different entities by different organizational departments (such as marketing, sales, and finance), which can lead to additional friction in the customer experience. This workshop seeks to bridge the gap between academic researchers and industrial practitioners who are interested in building holistic solutions for end-to-end customer journey optimization. In addition, with the rising popularity of generative AI and LLM, we want to use this venue to exchange ideas regarding their applications in different stages of customer journey, and how the new technologies could help businesses achieve their objectives.

This workshop introduces generative AI applications for enterprise, with a focus on retrieval-augmented generation (RAG) systems. Generative AI is a field of artificial intelligence that can create new content and solve complex problems. RAG systems are a novel generative AI technique that combines information retrieval with text generation to generate rich and diverse responses. RAG systems can leverage enterprise data, which is often specific, structured, and dynamic, to provide customized solutions for various domains. However, enterprise data also poses challenges such as scalability, security, and data quality. This workshop convenes researchers and practitioners to explore RAG and other generative AI systems in real-world enterprise scenarios, fostering knowledge exchange, collaboration, and identification of future directions. Relevant to the CIKM community, the workshop intersects with core areas of data science and machine learning, offering potential benefits across various domains.

Nowadays, while most machine learning research on customer journey optimization has focused on short-term success metrics such as click-through rates or optimal ad placement, there has been little consideration given to developing a coherent system for end-to-end customer journey optimization. Such a system would encompass all aspects of the customer experience, from presenting the right product value to the right users, to understanding a user's likelihood of conversion and long-term value to the platform, as well as their propensity for cross-selling and risk of churning. Currently, models and algorithms for customer journey optimization are often developed in isolation, leading to inefficiencies in modeling and data pipelines. Furthermore, the customer is often viewed as a collection of different entities by different organizational departments (such as marketing, sales, and finance), which can lead to additional friction in the customer experience. This workshop seeks to bridge the gap between academic researchers and industrial practitioners who are interested in building holistic solutions for end-to-end customer journey optimization. In addition, with the rising popularity of generative AI and LLM, we want to use this venue to exchange ideas regarding their applications in different stages of customer journey, and how the new technologies could help businesses achieve their KPIs.

Nowadays, while most machine learning research on customer journey optimization has focused on short-term success metrics such as click-through rates and optimal ad placement, there has been little consideration given to developing a coherent system for end-to-end customer journey optimization. Such a system would encompass all aspects of the customer experience, from presenting the right product value to the right users, to understanding a user's likelihood of conversion and long-term value to the platform, as well as their propensity for cross-selling and risk of churning. Currently, models and algorithms for customer journey optimization are often developed in isolation, leading to inefficiencies in modeling and data pipelines. Furthermore, the customer is often viewed as a collection of different entities by different organizational departments (such as marketing, sales, and finance), which can lead to additional friction in the customer experience. This workshop seeks to bridge the gap between academic researchers and industrial practitioners who are interested in building holistic solutions for end-to-end customer journey optimization. By fostering collaboration and cross-disciplinary discussion, the workshop aims to accelerate progress in this rapidly evolving field.

At present, most machine learning research on customer optimization focuses on short term success of the customers by addressing questions such as - which users have a higher propensity to click? Where to place one ad/multiple contents on a web page? What is the most appropriate time to show content? There has been less/little thought put into building a coherent system for the long term/end-end customer optimization from acquisition by understanding a user's propensity to convert to a particular product at a certain time, to user's ability to be successful long term on a platform as measured by CLV (Customer Lifetime Value), to users' ability to buy more products (cross sell) on the same platform, and finally users propensity to churn. Currently, such models and algorithms are built in isolation to serve a single purpose which leads to inefficiencies in modeling and data pipelines. Also, most of the time the customer is not looked at as a single entity - but each product/subgroup within an organization (marketing, sales, product growth, go-to-market, product) considers the customer independently. This workshop aims to connect academic researchers and industrial practitioners who are working on, or interested in building holistic systems and solutions in the field of end to end customer journey optimization.

The interpretability or explainability of AI systems (XAI) has been a topic gaining renewed attention in recent years across AI and HCI communities. Recent work has drawn attention to the emergent explainability requirements of in situ, applied projects, yet further exploratory work is needed to more fully understand this space. This paper investigates applied AI projects and reports on a qualitative interview study of individuals working on AI projects at a large technology and consulting company. Presenting an empirical understanding of the range of stakeholders in industrial AI projects, this paper also draws out the emergent explainability practices that arise as these projects unfold, highlighting the range of explanation audiences (who), as well as how their explainability needs evolve across the AI project lifecycle (when). We discuss the importance of adopting a sociotechnical lens in designing AI systems, noting how the “AI lifecycle” can serve as a design metaphor to further the XAI design field.

Chatbots are widely employed in various scenarios. However, given the high costs of chatbot development and chatbots’ tremendous social influence, chatbot failures may inevitably lead to a huge economic loss. Previous chatbot evaluation frameworks rely heavily on human evaluation, lending little support for automatic early-stage chatbot examination prior to deployment. To reduce the risk of potential loss, we propose a computational approach to extracting features and training models that make a priori prediction about chatbots’ popularity, which indicates chatbot general performance. The features we extract cover chatbot Intent, Conversation Flow, and Response Design. We studied 1050 customer service chatbots on one of the most popular chatbot service platforms. Our model achieves 77.36% prediction accuracy among very popular and very unpopular chatbots, making the first step towards computational feedback before chatbot deployment. Our evaluation results also reveal the key design features associated with chatbot popularity and offer guidance on chatbot design.

This lecture-style tutorial, which mixes in an interactive literature browsing component, is intended for the many researchers and practitioners working with text data and on applications of natural language processing (NLP) in data science and knowledge discovery. The focus of the tutorial is on the issues of transparency and interpretability as they relate to building models for text and their applications to knowledge discovery. As black-box models have gained popularity for a broad range of tasks in recent years, both the research and industry communities have begun developing new techniques to render them more transparent and interpretable. Reporting from an interdisciplinary team of social science, human-computer interaction (HCI), and NLP/knowledge management researchers, our tutorial has two components: an introduction to explainable AI (XAI) in the NLP domain and a review of the state-of-the-art research; and findings from a qualitative interview study of individuals working on real-world NLP projects as they are applied to various knowledge extraction and discovery at a large, multinational technology and consulting corporation. The first component will introduce core concepts related to explainability in NLP. Then, we will discuss explainability for NLP tasks and report on a systematic literature review of the state-of-the-art literature in AI, NLP and HCI conferences. The second component reports on our qualitative interview study, which identifies practical challenges and concerns that arise in real-world development projects that require the modeling and understanding of text data.

Chatbot has become an important solution to rapidly increasing customer care demands on social media in recent years. However, current work on chatbot for customer care ignores a key to impact user experience - tones. In this work, we create a novel tone-aware chatbot that generates toned responses to user requests on social media. We first conduct a formative research, in which the effects of tones are studied. Significant and various influences of different tones on user experience are uncovered in the study. With the knowledge of effects of tones, we design a deep learning based chatbot that takes tone information into account. We train our system on over 1.5 million real customer care conversations collected from Twitter. The evaluation reveals that our tone-aware chatbot generates as appropriate responses to user requests as human agents. More importantly, our chatbot is perceived to be even more empathetic than human agents.

Successful human interactions are based on becoming aware of other's emotion and making adaptations accordingly. However, understanding emotion is a complex task that has generated countless debates among researchers over the past decades. The abstract nature of human emotion highlights the need for a new data-driven approach that can better describe and compare across fine-grained emotional states. In this study, we propose Seemo, a novel neural embedding framework, which allows us to map human emotions into vector space representations. Seemo is trained using Twitter data and is evaluated on two fundamental use cases in traditional emotion research: determining the underlying dimensions of emotions and identifying the set of basic emotions. The evaluation reveals that on both tasks Seemo can generate results consistent with the mainstream theories. Results also show that the vector space representation of Seemo can effectively decode the important relationships between emotions that were usually not explicitly presented.

Users are rapidly turning to social media to request and receive customer service; however, a majority of these requests were not addressed timely or even not addressed at all. To overcome the problem, we create a new conversational system to automatically generate responses for users requests on social media. Our system is integrated with state-of-the-art deep learning techniques and is trained by nearly 1M Twitter conversations between users and agents from over 60 brands. The evaluation reveals that over 40% of the requests are emotional, and the system is about as good as human agents in showing empathy to help users cope with emotional situations. Results also show our system outperforms information retrieval system based on both human judgments and an automatic evaluation metric.

Emotion analysis of online customer service conservation is important for good user experience and customer satisfaction. However, conventional metrics do not fit this application scenario. In this work, by collecting and labeling online conversations of customer service on Twitter, we identify 8 new metrics, named as tones, to describe emotional information. To better interpret each tone, we extend the Latent Dirichlet Allocation (LDA) model to Tone LDA (T-LDA). In T-LDA, each latent topic is explicitly associated with one of three semantic categories, i.e., tone-related, domain-specific and auxiliary. By integrating tone label into learning, T-LDA can interfere the original unsupervised training process and thus is able to identify representative tone-related words. In evaluation, T-LDA shows better performance than baselines in predicting tone intensity. Also, a case study is conducted to analyze each tone via T-LDA output.

Crowdsourcing has proven to be an effective method for generating labeled data for a range of NLP tasks.However, multiple recent attempts of using crowdsourcing to generate gold-labeled training data for semantic role labeling (SRL) reported only modest results, indicating that SRL is perhaps too difficult a task to be effectively crowdsourced.In this paper, we postulate that while producing SRL annotation does require expert involvement in general, a large subset of SRL labeling tasks is in fact appropriate for the crowd.We present a novel workflow in which we employ a classifier to identify difficult annotation tasks and route each task either to experts or crowd workers according to their difficulties.Our experimental evaluation shows that the proposed approach reduces the workload for experts by over two-thirds, and thus significantly reduces the cost of producing SRL annotation at little loss in quality.

In this paper, we present computational models to predict Twitter users' attitude towards a specific brand through their personal and social characteristics. We also predict their likelihood of taking different actions based on their attitudes. In order to operationalize our research on users' attitude and actions, we collected ground-truth data through surveys of Twitter users. We have conducted experiments using two real world datasets to validate the effectiveness of our attitude and action prediction framework. Finally, we show how our models can be integrated with a visual analytics system for customer intervention.

With the recent advances in using data analytics to automatically infer one's personality traits from their social media data, users are facing a growing tension between the use of the technology to aid self development in workplace and the privacy concerns of such use. Given the richness of personality data that can be derived today and the varied sensitivity of revealing such data, it is a non-trivial task for users to configure their privacy settings for sharing and protecting their derived personality data. Here we present the design, development, and evaluation of an interactive visualization tool, VeilMe, which helps users configure the privacy settings for the use of their personality portraits derived from social media. Unlike other privacy configuration tools, our tool offers two distinct advantages. First, it presents a novel and intuitive visual interface that aids users in understanding and exploring their own personality traits derived from their social media data, and configuring their privacy preferences. Second, our tool helps users to jump start their privacy settings by suggesting initial sharing strategies based on a set of factors, including the users' personality and target audience. We have evaluated the use of our tool with 124 participants in an enterprise context. Our results show that VeilMe effectively supports various user privacy configuration tasks, and also suggest several design implications, including the approaches to personalized privacy configurations.

Crowd feedback systems offer designers an emerging approach for improving their designs, but there is little empirical evidence of the benefit of these systems. This paper reports the results of a study of using a crowd feedback system to iterate on visual designs. Users in an introductory visual design course created initial designs satisfying a design brief and received crowd feedback on the designs. Users revised the designs and the system was used to generate feedback again. This format enabled us to detect the changes between the initial and revised designs and how the feedback related to those changes. Further, we analyzed the value of crowd feedback by comparing it with expert evaluation and feedback generated via free-form prompts. Results showed that the crowd feedback system prompted deep and cosmetic changes and led to improved designs, the crowd recognized the design improvements, and structured workflows generated more interpretative, diverse and critical feedback than free-form prompts.

Hundreds of thousands of crowdfunding campaigns have been launched, but more than half of them have failed. To better understand the factors affecting campaign outcomes, this paper targets the content and usage patterns of project updates -- communications intended to keep potential funders aware of a campaign's progress. We analyzed the content and usage patterns of a large corpus of project updates on Kickstarter, one of the largest crowdfunding platforms. Using semantic analysis techniques, we derived a taxonomy of the types of project updates created during campaigns, and found discrepancies between the design intent of a project update and the various uses in practice (e.g. social promotion). The analysis also showed that specific uses of updates had stronger associations with campaign success than the project's description. Design implications were formulated from the results to help designers better support various uses of updates in crowdfunding campaigns.

Feedback on designs is critical for helping users iterate toward effective solutions. This paper presents Voyant, a novel system giving users access to a non-expert crowd to receive perception-oriented feedback on their designs from a selected audience. Based on a formative study, the system generates the elements seen in a design, the order in which elements are noticed, impressions formed when the design is first viewed, and interpretation of the design relative to guidelines in the domain and the user's stated goals. An evaluation of the system was conducted with users and their designs. Users reported the feedback about impressions and interpretation of their goals was most helpful, though the other feedback types were also valued. Users found the coordinated views in Voyant useful for analyzing relations between the crowd's perception of a design and the visual elements within it. The cost of generating the feedback was considered a reasonable tradeoff for not having to organize critiques or interrupt peers.

Online communities are important in enterprises, helping workers to build skills and collaborate. Despite their unique and critical role fostering successful communities, community leaders have little direct support in existing technologies. We introduce CommunityCompare, an interactive visual analytic system to enable leaders to make sense of their community's activity with comparisons. Composed of a parallel coordinates plot, various control widgets, and a preview of example posts from communities, the system supports comparisons with hundreds of related communities on multiple metrics and the ability to learn by example. We motivate and inform the system design with formative interviews of community leaders. From additional interviews, a field deployment, and surveys of leaders, we show how the system enabled leaders to assess community performance in the context of other comparable communities, learn about community dynamics through data exploration, and identify examples of top performing communities from which to learn. We conclude by discussing how our system and design lessons generalize.

Affect influences workplace collaboration and thereby impacts a workplace's productivity. Participants in face-to-face interactions have many cues to each other's affect, but work is increasingly carried out via computer-mediated channels that lack many of these cues. Current presence systems enable users to estimate the availability of other users, but not their affective states or communication preferences. This work demonstrates the feasibility of estimating affective state and communication preferences from a stream of presence states that are already being shared in a deployed presence system.

Critique is an indispensible part of creative work and many online communities have formed for this shared purpose. As design choices within the communities can impact the effectiveness of the critiques produced, it is important to study these communities and offer guidance for decisions. In this paper, we report the results of a case study exploring one large online community dedicated to critique in the domain of digital photography. We analyzed a large corpus of interaction data to understand the benefit of participation, the response dynamics, factors predicting critique ratings, and patterns of reciprocal interaction. Interviews with users were also conducted to uncover motives for participation and expectations of the critiques within the community. The results and insights gained from this work were distilled into recommendations for improving the design of systems that support community-based critique of creative artifacts.

Idea pipelines enable open innovation within organizations but require the evaluation teams to assess large numbers of ideas. To help filter promising ideas, community voting is often included as part of the pipeline but the outcome of the voting rarely aligns with the ideas selected by the team. To address this problem, we introduce a new scoring model for increasing the findability of promising ideas within idea pipelines. In the model, each participant need only score a subset of the ideas, ideas are scored independently, and the individual scores can be aggregated. We tested the model on an authentic data set and found our model filters ideas chosen by an evaluation team better than community votes.

Generating and managing multiple ideas is a fundamental part of the creative process for both individual and teams of designers. To be useful for early design, computer tools must effectively support this common practice. This paper proposes a demonstration of a new collaborative interface for managing ideas in computer-based design tools. The core of the interface provides interactive spatial maps for creating, organizing, and reflecting on ideas. The interface was also substantially revised in response to lessons learned from a study comparing its use to the use of other tools for managing multiple ideas during early design work. The revised interface allows designers to tag and filter the idea space, arrange ideas in three distinct views, and efficiently compose and decompose content from multiple ideas.

We present a new feature extraction method, which called the complete two-dimensional principal component analysis (Complete 2DPCA), for image registration. Complete 2DPCA is based on 2D image matrices. Two image covariance matrices are constructed directly using the original image matrix and their eigenvectors are derived for image feature extraction. In the 2D image registration scheme, we propose complete 2DPCA to extract features from the image sets, and these features are input vectors of feedforward neural networks (FNN). Neural network outputs are registration parameters with respect to reference and observed image sets. Comparative experiments are performed between complete 2DPCA based method and other feature based methods. The results show that the proposed method has an encouraging performance.

We present a comprehensive system for weather data visualization. Weather data are multivariate and contain vector fields formed by wind speed and direction. Several well-established visualization techniques such as parallel coordinates and polar systems are integrated into our system. We also develop various novel methods, including circular pixel bar charts embedded into polar systems, enhanced parallel coordinates with S-shape axis, and weighted complete graphs. Our system was used to analyze the air pollution problem in Hong Kong and some interesting patterns have been found.

In this paper, a novel image registration method is proposed. In the proposed method, kernel independent component analysis (KICA) is applied to extract features from the image sets, and these features are input vectors of feedforward neural networks (FNN). Neural network outputs are those translation, rotation and scaling parameters with respect to reference and observed image sets. Comparative experiments are performed between KICA based method and other six feature extraction based method: principal component analysis (PCA), independent component analysis (ICA), kernel principal component analysis (KPCA), the discrete cosine transform (DCT), Zernike moment and the complete isometric mapping (Isomap). The results show that the proposed method is much improved not only at accuracy but also remarkably at robust to noise.

In the literature of astronomical spectral classification, linear principle component analysis (PCA) was frequently employed to extract features of spectra data. However, the spectral data are too complicated to be well described by a linear model. In this paper, kernel independent component analysis (KICA), which contains a nonlinear kernel mapping component, is adopted to extract features from the spectra of galaxies. Then, a radial basis function neural network is adopted as a classifier to implement the classification. Experiments with real-world spectral data set show that KICA is a very appropriate technique to describe the important features of celestial objects, and the correct classification rate is improved compared with PCA method.