AI requires interdisciplinary teams, Quality Data & Explainability

Data Transformers Podcast AI requires interdisciplinary teams, Quality Data & Explainability 1x 00:00 / 00:24:34 Subscribe Share Apple Podcasts Google Podcasts Spotify Stitcher TuneIn RSS Feed Share Link Embed

AI requires interdisciplinary teams, Quality Data & Explainability

' /> Apple Podcasts Google Podcasts Spotify Stitcher https://youtu.be/yCxofWqPyyU Episode Title : AI requires interdisciplinary teams, Quality Data & Explainability Episode Summary: Artificial Intelligence and Machine Learning projects require interdisciplinary skills in devops, SW engineering in addition to hard core data science coding skills. Additionally, lot of rigor needs to be put into cleaning up the data that is fed into the models. On an interesting note, AI models can also be used for improving data quality as well. Lastly, Explainability of models and data is becoming important and as such explainability needs to be baked in.

Topics discussed in this episode:

Leveraging past experience in current job (01:47): Fiona’s past experience in SW engineering, Ph.D., and SW development has been very helpful in the current job as the head of AI. The rigor associated with SW Engg, the discipline of research methodology is extremely helpful in the multidisciplinary area of data science and AI. Deploying models (05:15): Deploying machine learning models requires dev ops, infra skillset, testing strategies, SW versioning of not only code but also data etc. So the data science teams need inter-disciplinary skills like dev ops, data quality, data governance, sw engineering etc. in addition to hard code coding. AI for improving data quality (07:07): AI requires quality data. But AI can also be used for improving data quality. As the data is fed in, AI models can be used to identify data anomalies and outliers and that information can be used to prescribe data quality tasks. While matching and de-duplicating, the models can use deterministic methods to identify anomalies and then use human in the middle to resolve the low confidence areas. Explainability AI ground-up (14:13): Explainability and transparency in machine learning models has gained a lot of traction. In real-life cases, an example of a doctor prescribing a treatment predicted by a model but unable to explain the reasoning would not work. There should be transparency around representative data and the models used. So it is important to build that transparency and explainability from ground up. Resources mentioned in this episode: https://Datactics.Com  Call To Action: Connect with Fiona Browne on LinkedIn:  https://www.linkedin.com/in/flbrowne/ Data Transformers Podcast

Listen Now!

Join Peggy and Ramesh as they explore the exciting world of Data Management, Data Analytics, Data Governance, Data Privacy, Data Security, Artificial Intelligence, Cloud Computing, Internet Of Things. Apple Podcasts Google Podcasts Spotify Stitcher