ISG Provider Lens™ Intelligent Automation – Platforms and Products - Process Discovery and Mining - U.S. 2023

19 Dec 2023
by Ashwin Gaidhani, Mukesh Ranjan, Sameen Mohammed Siddique, Jan Erik Aase
$2499

Automation, augmentation and autonomy are consolidating to enhance business productivity

Changing market dynamics drive the functional architecture design, including intelligent capabilities and human augmentation, to amplify performance. The performance component has expanded beyond speed and incorporates efficiency, experience and economics, directly impacting all value chain stakeholders. With the
introduction of emerging technologies, humanmachine augmentation can seamlessly manage more complex activities than ever before. The independent software vendors (ISVs) market is evolving with the emergence of platforms that connect and integrate solutions, capabilities and products to develop offerings specific to an industry, persona and function. This drives innovation to introduce autonomy and intelligent orchestration that helps automation drive changes to the next level. Once there is a clear vision and path for automation,
augmentation and autonomy, businesses aim to extract intelligence and build visualization panes that present actionable information. This enterprise intelligence visualization practice is driven by analytics and data sciences converging with ML on a platform-like offering to drive deep business insights.

Designing intelligent conversation is being explored to elevate and enhance user experience

Conversational AI (ConAI) is evolving to integrate multiple modes of communication, including text, voice and visual inputs. Generative AI (GenAI) models enhance ConAI systems by enabling a more natural and versatile
interaction. For example, AI - powered chatbots like Ada Health incorporate multimodal capabilities to understand and respond to textual descriptions and images in healthcare. GenAI is revolutionizing ConAI platforms by enabling them to understand and respond in a highly personalized manner. These platforms
can grasp user preferences and contextual cues through advanced natural language processing (NLP) and ML, providing more tailored and relevant responses. In the retail sector, virtual assistants like those used by global e-commerce leaders leverage generative models to analyze user behavior and offer personalized product recommendations, creating a more engaging shopping experience and ambiance.

Conversational agents are becoming more sophisticated, mimicking human-like conversational styles and nuances. The recently popular transformers and large language models (LLMs) have demonstrated the ability to generate contextually relevant and coherent responses, contributing to more natural and engaging interactions, while also offering practicable controls, such as temperature and p-square values, to tune the model outputs based on requirements of creative responses but avoiding potential AI hallucinations. Prompt engineering and low-code/no-code (LCNC) features enable casual and enterprise business users to leverage GenAI in conversations and day-to-day work such as proposal drafting, semantic searches and Q&A from large documents, etc. These trends are particularly evident in the customer service domain, where virtual assistants like Google’s Duplex can make reservations or schedule appointments with human- like conversational experiences, which extend the discussions on the real-world availability of artificial general intelligence (AGI).

Semantically and contextually smart translation features in AI models powered by advanced GenAI have broken language barriers in conversational applications by providing robust cross-lingual capabilities. These models can understand and generate content in multiple languages, enhancing accessibility and inclusivity. This is crucial for customer support systems to efficiently handle inquiries in various languages in global business operations. For instance, companies like Microsoft and Google are incorporating GenAI into their  translation and language understanding services to enable seamless multilingual interactions, augmenting ConAI platforms with contextual emotional and limited social intelligence and allowing them to understand and respond to user emotions and interactions. By incorporating sentiment analysis, these systems can adapt their tone and content based on the user’s mood. In social media monitoring, tools utilize generative models to analyze and respond to user comments with a nuanced understanding of sentiment, helping businesses manage their online reputation and brand equity. 

Document definition and information extraction are evolving at the pace of technology.

Document processing solutions involve advanced OCR technology, which enables the extraction of information from documents with greater accuracy. ABBYY and UiPath are leading vendors in this space. For example, in the insurance sector, UiPath’s Document Understanding uses OCR to extract data from insurance claims, streamlining the claims processing workflow. Vendors like ABBYY and Tesseract are pushing the boundaries of OCR, making it possible to extract text from complex documents, handwritten text and even low-quality scans. ABBYY’s FineReader extracts and analyzes text from legal documents, streamlining contract management processes in the legal sector. In document processing, NLP systems help understand the  context and meaning of text within documents.

Leveraging NLP, platforms like IBM Watson Discovery can extract insights from unstructured data. In the healthcare industry, semantic content discovery solutions are applied to extract relevant information from medical records, enhancing the efficiency of data analysis in patient care. Document Understanding platform, for instance, uses AI to classify and extract data from documents, allowing businesses to automate tasks such as invoice processing. UiPath’s technology helps organizations streamline the accounts payable processes in the finance sector. With data privacy regulations becoming more stringent, document processing platforms also focus on data security and compliance. Integrated IDP solutions offer secure cloud-based document management and compliance solutions. Specific solutions ensure patient records are handled in compliance with the Health Insurance Portability and Accountability Act (HIPAA). GPT-3 and BERT are being integrated into document processing platforms to facilitate content generation and summarization. In the content marketing sector, AI technology stacks, including transfer models, generate blog posts, articles and marketing copy, reducing the time and effort required for content creation. NLU is a key trend in automated IDP, enabling systems to understand the context and meaning of text within documents. Leveraging NLU platforms for document processing can help in improved document interpretation. In  insurance claims processing, IDP platforms are employed to extract relevant information from medical reports, facilitating quicker and more accurate claims assessments.

Process mining space is consolidating to embed as a core business process function

Traditional process mining tools primarily focus on retrospective analysis of historical data, while  contemporary solutions leverage advanced ML algorithms for real-time monitoring and predictive insights. These tools identify inefficiencies and provide proactive recommendations for process optimization. Process discovery solutions leverage advanced analytics and automation. For example, modern tools leverage ML algorithms to analyze large datasets and identify hidden patterns in business processes. They increasingly focus on real-time process monitoring and event log analysis to provide organizations with insights into their operations.

Additionally, there is a growing emphasis on user-friendly interfaces and visualization techniques to enhance the accessibility of process mining for non-technical users. This involves integrating collaborative technologies and human-centric approaches. Tools are now emphasizing the involvement of stakeholders at various stages of the design process, fostering a more inclusive, persona-based, and innovative environment to support users at all levels and tech proficiencies. Design thinking principles are incorporated into process design tools, emphasizing empathy, ideation and prototyping. Moreover, the adoption of the LCNC platform is increasing, enabling business users to actively participate in the design and innovation of processes without extensive coding knowledge. Process simulation technologies are also evolving to offer more accurate and dynamic representations of real-world scenarios. Simulation tools now integrate with real-time data streams, allowing organizations to model and analyze processes in  complex, dynamic environments. The integration of GenAI in simulation tools is a notable trend, enabling the creation of alternative process scenarios and optimizing workflows.

Further, GenAI algorithms can suggest innovative process variations, allowing organizations to explore new possibilities and enhance resource allocation and risk management decision-making. One key trend is the integration of ML models to discover and map intricate business processes automatically. These models analyze vast datasets, identifying real-time patterns, deviations and bottlenecks, enabling organizations to respond swiftly to changing market conditions and continuously improve their operations. These enhancements democratize process mining, making it accessible to a broader audience within organizations, including business analysts and managers.

In the field of automation, consolidation primarily occurs at the core architecture level, while differentiation takes place at the consumption layer. Here, various capabilities have branched out as independent solutions, expanding within the same domain. Integration platforms are being developed to converge and complement individual capabilities to solve many technological challenges. Enterprise clients are transforming their business models, processes and outcomes based on solution maturity and capability, and the transformation is irreversible. ISVs have aligned their solutions with specific industries, inorganically  integrating the capabilities to enhance their growth and adoption rates. This approach has enabled them to expand beyond their traditional areas of influence, now offering enterprise-wide solutions and provisioning capabilities.

Access to the full report requires a subscription to ISG Research. Please contact us for subscription inquiries.

Page Count: 34
QUESTIONS?
To purchase this product or for more information, please contact your account manager: