Learning & Innovation Research Manager| Project Management Institute (PMI)Spain
Data quality and quantity is particularly important as we think about leveraging AI on projects. Considerations include the diversity and comprehensiveness of the data that is available to us.
Have you ever encountered unexpected challenges or pitfalls while using data in your projects? How did you navigate the situation and find a resolution? Saving Changes...
Omar JabbarProject Management and Digital Transformation Consultant| OGreen IT Service Inc.Ontario, Canada
Certainly, unexpected challenges arise frequently when working with data. In such instances, I rely on a structured approach involving thorough analysis, iterative problem-solving, and collaboration.
I assess the issue, gather relevant information, and consult with domain experts. Then, I employ data cleaning, feature engineering, and model tuning techniques as needed. Through continuous validation and testing, I refine my approach until achieving satisfactory results.
Documentation ensures reproducibility, and staying open to learning new methods helps overcome obstacles. Ultimately, by remaining adaptable and persistent. Saving Changes...
In one of our recent project the main problems we faced when we were working with large amount of data were inaccuracies, inconsistency and duplicated or outdated data and most importantly missing data . This not only forced us to put sufficient extra time to first understand the data and then prepare it for further use. Then again when we used our model to predict the data based on previous data ,it took us sufficient time to achieve a respectable accuracy . And all this extra effort was due to missing and inaccurate data. However we managed it after our sustained effort as this exercise gave insights on what are the major pitfalls in current system and we suggest measures to overcome it . At the end we were successful in implementing correct data management practices and our prediction was also improved significantly with time. Saving Changes...
Leonard MarcheseFounder & President| Rethink Inc.Chicago, Il, United States
This may seem opposite from the intention of the question related to AI projects, but I find that withholding of data may create an information gap with negative impact. Effectively collaborating and communicating with stakeholders helps mitigate this risk, as does ensuring you have included all the right contributors, diverse perspectives, and relevant observations. Whether the data is in AI or not it mut be reliable, relevant, cohesive, and complete. Saving Changes...
I lead several data analytics projects in my career. I can say that It is crucial that all stakeholders are aware of where those data come from (data sources), timing for data capture, and the rules applied during the trasformation process before the data is loaded into the system for the purpose. Data lineage is the technique that drives users to have a common understanding of data meaning and analytics results.
Data Governance is the the discipline that should be always put in place before starting a project which outcomes are driven by data. Saving Changes...
Anonymous
In our projects, we develop set checklists to establish the sequence as well as the validity of data that should be considered by the model. PM takes the help of data scientist to monitor the data quality and reject unrelated data Saving Changes...
In my personal opinion, any and all data used for any activity must be checked and validated before it gets used. Even more important to follow the above especially when the data itself or their analyzed outputs/ trends etc are being used for decision making.
Garbabe in results in garbabe out and can be the reason for a fallen project. Saving Changes...
I often see organizations or businesses failing to understand the data needs, which should be understood and acted upon by the application you are planning to develop. For example, if you are planning to develop a healthcare solution which works for all ages and genders, the training and evaluation dataset should obviously include such variability.
Once the data is available to train your models, make sure you visualize datasets, which will give you a better understanding on quality and data augmentation needs. Such quality checks before training models could also save compute resources and carbon emissions by blindly running trainings and realizing later that the input data quality was not good enough.
In addition, different countries have specific data regulations (even though you are using open source datasets), which should be taken care of before you even see the first images. You, as a project manager should be aware of standard data practices and train your team accordingly. Saving Changes...
Serge Ateba, PMPDirector Project Controls| KBRHouston, United States
Jan 13, 2024 3:47 AM
Replying to Nikita Jha
...
In my previous project, the objective was to reduce obsolescence for database technologies across the globe aka many regions. The timeframe for communication with immense amount of data took two weeks cycle for response and plan of action. However the data kept growing as many assets were not listed in a particular region as API failed to retrieve information at the right time. Hence I took that region as a subset project with dedicated resources and fast tracked to bring the region at par with other regions. It was immensely challenging but at the end it was rewarding in terms of achievement and stakeholders appreciation.
Hi Nikita, thank you for you contribution. I came across a similar situation for an Oil & gas company. In the case of legacy systems where data reside in data warehouses or data centers, this will increase complexity. Now with solutions from vendors like AWS, Azure or GCP, we can leverage them and minimize the risk Saving Changes...
Project managers must ensure that quality of Data used to train AI model in any Gen AI related project is not compromised because they are accountable for the outcome of the information generated from such AI model Saving Changes...
Ranadeer SappidiProject Manager| Kerv Digitalhyderabad, TG, India
In my current project, We are in situation where customer had raised a question on how to minimize the Data storage on Cloud Platform since the customer is from Pharma industry and from Europe based got lot of GDPR policies and they need to comply the same. This course from PMI on Data Landscape of GenAI for Project Manager is very much helping me to execute my project and define the Data Governance Model. Saving Changes...
PMI Team Member
