BEGIN:VCALENDAR PRODID:-//github.com/ical-org/ical.net//NONSGML ical.net 5.1.4//EN VERSION:2.0 BEGIN:VEVENT DESCRIPTION:Traditional Voice of Customer (VOC) data analysis methods are l abor-intensive and time-consuming\, often requiring hundreds of hours for  qualitative analysis. This discussion introduces an AI-powered workflow based on commercially available large language models that automates the m ost laborious steps of VOC qualitative data analysis\, reducing processing time from hundreds of hours to minutes while maintaining acceptable accur acy and improving consistency. The proposed workflow leverages the generic GPT-4o-Mini model to process and prepare transcript data\, identify custo mer need statements\, and organize these statements into a hierarchical st ructure for further quantitative analysis. The workflow includes stages su ch as transcript preparation\, chunking text for analysis\, identifying an d scoring needs\, and clustering needs into prioritized categories. Evalua tion results demonstrate a significant reduction in time and effort\, with the AI workflow achieving a 99% reduction in analysis time compared to tr aditional methods with sufficient specificity of customer need if the chun k sizes are adequately small. Furthermore\, the results help to overcome s everal typical cognitive and motivational biases of the analysts. The find ings highlight the potential of AI to enhance the efficiency and consisten cy of VOC analysis\, providing valuable insights for product development a nd customer satisfaction improvement. While the discussion will mainly be presentation and Q&A\, we will try to open it up in general for lessons le arned in using AI for qualitative analysis of documents. At the end\, and if there is time\, we will discuss the current work building on these less ons-learned. We have set up AI workflows on generic LLMs for semi-automati ng threat modeling to develop security requirements for custom systems. T his event is brought to you by the VA AWWA | VWEA Technology Services Comm ittee. There is no cost\, and you don't need to register.  For Zoom conne ction information\, please email Cynthia.Barnes@VaAWWA.org Please note tha t this session will not be recorded or transcribed in order to encourage a ctive conversation and participation. DTEND:20250417T170000Z DTSTAMP:20260607T051525Z DTSTART:20250417T160000Z LOCATION:Virtual Zoom SEQUENCE:1 STATUS:CONFIRMED SUMMARY:Roundtable Discussion: Automating Voice of Customer Analysis with A I Workflows Built on Generic GPT-40-Mini Model UID:13e80ff2-fd13-4c92-9ff9-0013deee90b7-638805024000000000 X-ALT-DESC;FMTTYPE=text/html:
\n

< span style="font-family:tahoma\,geneva\,sans-serif">Traditional Vo ice of Customer (VOC) data analysis methods are labor-intensive and time-c onsuming\, often requiring hundreds of hours for \;qualitative analysi s.

\n\n

This discussion introduces an AI-powered workflow based on commercially available large language models that automates the most laborious steps of VOC qualitative data analysis\, reducing processing time from hundreds of hours to minutes while maintain ing acceptable accuracy and improving consistency. The proposed workflow l everages the generic GPT-4o-Mini model to process and prepare transcript d ata\, identify customer need statements\, and organize these statements in to a hierarchical structure for further quantitative analysis. The workflo w includes stages such as transcript preparation\, chunking text for analy sis\, identifying and scoring needs\, and clustering needs into prioritize d categories. Evaluation results demonstrate a significant reduction in ti me and effort\, with the AI workflow achieving a 99% reduction in analysis time compared to traditional methods with sufficient specificity of custo mer need if the chunk sizes are adequately small. Furthermore\, the result s help to overcome several typical cognitive and motivational biases of th e analysts. The findings highlight the potential of AI to enhance the effi ciency and consistency of VOC analysis\, providing valuable insights for p roduct development and customer satisfaction improvement.

\n\n\n
\n\n
This event is brought to you by the VA AWWA | VWEA Technology Services Committee. There is no cost\, a nd you don't need to register. \; For Zoom connection information\, pl ease email Cynt hia.Barnes@VaAWWA.org
\n\n
\n
\n\n\n
\n

\n
\n
\n
\n END:VEVENT END:VCALENDAR