“It’s much easier to get to a good prototype if we know what the inputs to a prototype are, which include data about the people who are going to use the solution, their usage scenarios, use cases, attitudes, beliefs…all these kinds of things.” - Brian O’Neill ⬆️ Join 2k readers: tap “Visit my website” to get my latest thinking in your inbox 🚫 Don’t like LinkedIn’s algo deciding what you see? Tap the bell under my profile banner to get all my posts ♻️ Please repost if this was useful 🎧: https://bit.ly/3WXsxKT
Brian T. O'Neill’s Post
More Relevant Posts
-
“It’s much easier to get to a good prototype if we know what the inputs to a prototype are, which include data about the people who are going to use the solution, their usage scenarios, use cases, attitudes, beliefs…all these kinds of things.” - Brian O’Neill ⬆️ Join 2k readers: tap “Visit my website” to get my latest thinking in your inbox 🚫 Don’t like LinkedIn’s algo deciding what you see? Tap the bell under my profile banner to get all my posts ♻️ Please repost if this was useful 🎧: https://bit.ly/3WXsxKT
144 – The Data Product Debate: Essential Tech or Excessive Effort? with Shashank Garg, CEO of Infocepts (Promoted Episode) | Designing for Analytics (Brian T. O'Neill)
https://designingforanalytics.com
To view or add a comment, sign in
-
Very interesting article.
Beyond Dashboards: The Future Of Analytics And Business Intelligence?
social-www.forbes.com
To view or add a comment, sign in
-
🌟 Time Series Forecasting: A Deep Dive into Predictive Analytics 🌟 I am thrilled to share my latest project on Time Series Forecasting, where I explored advanced techniques to predict future trends using historical data. Key Highlights of the Project 1️⃣ Data Understanding and Preparation: - Worked with a rich time-series dataset, transforming raw data into a structured format suitable for analysis. - Addressed missing values, outliers, and inconsistencies to ensure data quality. - Extracted key features like trend, seasonality, and cyclic patterns to gain an initial understanding of the dataset. 2️⃣ Exploratory Data Analysis (EDA): - Visualized time series data using advanced plotting techniques to uncover trends, seasonality, and anomalies. - Used statistical tools to understand autocorrelation and stationarity, critical for selecting suitable forecasting models. 3️⃣ Model Development and Implementation: - Built and compared multiple forecasting models: Statistical Models: ARIMA, SARIMA, and Exponential Smoothing for capturing linear patterns. - Machine Learning Models: LSTM and GRU networks for handling non-linear trends and long-term dependencies. - Applied feature scaling, sliding windows, and lag-based feature engineering for improved model input. - Tuned hyperparameters and validated models using techniques like grid search and time series cross-validation. 4️⃣ Evaluation and Performance: - Evaluated models using metrics such as MAE, MSE, and RMSE for accuracy and reliability. - Benchmarked traditional statistical models against modern deep learning approaches, identifying their strengths and limitations. 5️⃣ Results and Insights: - Developed a highly accurate and scalable forecasting framework adaptable to various domains like sales, demand, inventory, and financial projections. - Discovered actionable trends and seasonal patterns that could be directly implemented for strategic decision-making. Skills and Tools Used 1. Languages & Libraries: Python, Statsmodels, TensorFlow, Keras, Pandas, NumPy, Matplotlib, and Seaborn. 2. Techniques: Time series decomposition, feature engineering, model tuning, and cross-validation. 3. Concepts: Trend analysis, seasonality extraction, stationarity testing, and long-term dependency modeling. Impact This project not only enhanced my expertise in time series forecasting and predictive analytics but also demonstrated the power of data in driving informed decisions. It’s a step closer to bridging the gap between historical data and future insights. I’m eager to apply these skills to real-world challenges and collaborate on innovative projects in data science and analytics. 🚀 GitHub Link: https://lnkd.in/dwBYGWvP #DataScience #TimeSeriesForecasting #PredictiveAnalytics #MachineLearning #Python #DataDrivenDecisions #Innovation #DataVisualization
MindShift-Technologies-Task/Time_Series_Forecasting (1).ipynb at main · anamitra-22/MindShift-Technologies-Task
github.com
To view or add a comment, sign in
-
https://lnkd.in/gm_4tsni I'm thrilled to share the latest entry on my blog where I delve into topics that intrigue and inspire. Whether you're looking for something thought-provoking or just a quick read to spark creativity, I've got you covered. Check out my new post and let's connect through ideas and discussions. Looking forward to your thoughts and feedback!
Introduction to Data Analyst
gawaredipalee.blogspot.com
To view or add a comment, sign in
-
Insightful thought piece on a badly over-chewed topic. There's no dearth of literature devoted to PMF, but this one is delightfully precise and purposeful. Very 'Meta' in its approach and appeal. Call it a fleeting 'feeling' or deem it a 'broken' cause, PMF is such a key concept (even for software products) and contrary to popular perception, there's always scope for the twain to meet: the board room folks betting on the product, and the lab guys building it. If both groups approach PMF as they should, they surely inch closer to those dream tradeoffs. Dr. (Inv) Prof Vishal U S Rao you may find this paper interesting - refer our midnight conversations on MVP 🤣 Dmitry Vostokov 🇮🇪 Luca Bellonda Makarand Amte Bhaskar Roy Dr. Amol Bhikane Dr. B S Ajaikumar Naveen Kumar Amar Ambani Hitesh Jain Pritesh Mehta
Analytics and Product-Market Fit
medium.com
To view or add a comment, sign in
-
Just finished the course “Recommendation Systems: A Practical Introduction” by Miguel González-Fierro! Check it out: https://lnkd.in/dmdq7hsX #recommendersystems.
Certificate of Completion
linkedin.com
To view or add a comment, sign in
-
Digging Deeper: A Guide to Extracting Actionable Insights from User Feedback https://lnkd.in/dbntAkKu
Digging Deeper: A Guide to Extracting Actionable Insights from User Feedback
medium.com
To view or add a comment, sign in
-
Observability is an important topic in the industry with many companies competing to be your top pick. I wrote about the observability companies I'm watching in 2024. https://lnkd.in/eiMjC3Gx
Matthew Sanabria - Observability Companies to Watch in 2024
matthewsanabria.dev
To view or add a comment, sign in
-
Probably a good time to announce that I have been consulting for different ML startups for a while. My expertise is going from product to finetuning LLMs and back. Having done these end to end loops for various companies over a decade has some advantages :) See what we can do for you https://thevertex.ai/ Some conversations are exploratory, but we tie it to an actual launch to keep us motivated. More launches, less thought leadership.
PSA: Don’t be afraid to put your draft work out in public, friends! Publishing our early video search demo has gotten me: 1. Progress on a deal with one of the top 10 public companies 2. Feedback from some of the smartest data science folks in the world like Julia Turc, Saurabh Bhatnagar, Sidharth Ramachandran, Vik K., Jae Lee Publishing draft work means risking looking silly in public. And that's hard. But just do it. 💪🔥😊
Philipp Tsipman on LinkedIn: Video AI is coming! So video search will be critical. I want to share a…
linkedin.com
To view or add a comment, sign in
-
Announcing cStructure's Public Beta: After 18 months of development, I'm pleased to introduce cStructure - a collaborative platform for causal inference. Last week we launched in the US, this week are expanding to Canada. The Challenge: While leading data science teams, I kept hitting the same wall: causal analyses would fall apart in stakeholder meetings. Subject matter experts would ask why their favorite variable wasn't in the analysis, data scientists would latch on to correlation-based approaches (like XGBoost and SHAP**), and leadership tried to triangulate the competing voices. What should have been evidence-based decisions became gridlock. The Solution: Everything changed when we started using causal diagrams. These visual maps of cause-and-effect became a shared language between experts, analysts, and decision-makers. Domain knowledge could be captured precisely. Assumptions became explicit. Teams could focus on the right questions and controls. However, building and validating these models was painful, scattered across whiteboards, papers, and custom code. We built cStructure to make rigorous causal inference collaborative and accessible. The platform that empowers teams to: - Kickstart and revise a causal diagram with AI assistance - Build and validate causal models collaboratively - Detect potential biases automatically - Connect causal graphs directly to data - Run analyses in-browser or export to preferred environments Try it: https://cstructure.dev Features: https://lnkd.in/gtC4pikk Our team brings deep experience from life sciences, energy, and technology sectors, where we've seen these methods scale from startups to major healthcare systems, with experimental sample sizes from dozens to millions. We don't just provide tools — we also provide partnership in helping teams understand and apply causal methods. We're working on federated privacy-preserving learning, FAIR causal models, and integration with knowledge graphs to further empower teams in deciphering the world's complexity together. We're actively seeking feedback from: - Leaders looking to make well-informed decisions - Researchers working with causal methods - Data science teams building decision support systems - Domain experts interested in rigorous causal analysis #CausalInference #DataScience #Analytics #DecisionScience **https://lnkd.in/gePBdcbA...
Causality. Simplified.
cstructure.dev
To view or add a comment, sign in