Time series analysis with knime
0 likes234 views
Time series analysis allows Data Scientists to recognize trends, seasonality, and correlations within past data related to an organization to make predictions on which business decisions are based. Let’s take a look at how various industries use Time series analysis to make crucial decisions. ~ The airline industry can optimize travel routes by predicting future weather patterns, seasonal demands, or unexpected events. ~ Stockbrokers use it to predict correlations within stocks & market conditions to decide where to invest. ~ Supply Chain companies can predict weather conditions, traffic patterns, expected delivery times to optimize routes. But did you know that you can build Time Series models with minimum knowledge of coding? The KNIME Analytics Platform can make this happen. It uses a Graphical User Interface to allow Data Scientists who are just starting out and do not have extensive experience in coding. In this webinar, our Machine Learning expert will help you build time series models using the KNIME Analytics Platform. Business leaders and Data scientists must not miss this opportunity to arrive at smart, data-driven business decisions with the help of this platform.
![[WSO2Con USA 2018] Winning Strategy For Enterprise Integration to Empower Dig...](https://cdn.slidesharecdn.com/ss_thumbnails/1130kingpaul-180718055022-thumbnail.jpg?width=560&fit=bounds)
Ad
More Related Content
Similar to Time series analysis with knime (20)
More from Knoldus Inc. (20)
Ad
Recently uploaded (20)
Ad
Time series analysis with knime
- 1. Time Series Analysis with KNIME Presented By: Shubham Goyal Data Scientist Knoldus Inc. & MachineX Intelligence
- 3. 3 Our Agenda 01 Importance of supply chain management 02 Introduction to time series analysis 03 Components of time series (Autocorrelation, Seasonality, Stationarity) 04 Modeling time series 05 Knoldus Forecasting platform
- 4. 4 About Knoldus MachineX MachineX is a group of data wizards. We are a team of Data Scientist and engineers with a product mindset who deliver competitive business advantage.
- 5. 3 Pillars of Knoldus
- 6. Our Global Presence 8+ Years Years of Profitable Growth 155+ People Largest Scala + Spark + Tensorflow + Pytorch Services Company 04 Offices Offices globally 17+ Customers Multi-year Global Customers
- 7. Our Partners Through our strategic partnerships, we have an unwavering commitment to equip your organization with the knowledge, skills, expertise, resources and tools to succeed.
- 8. Knoldus MachineX Offerings Natural Language Processing Computer Vision Solutions Data mining Chatbot Development Artificial intelligence research and solutions
- 9. 9 An Intelligent Meeting Assistant Application Record Videos View DashBoard
- 10. 10
- 11. 11 Machine learning library in scala KSAI FishEye
- 12. Case Studies Enabling Intelligent systems for enterprises
- 13. Challenges: ● Forecasting earthquakes is one of the most important problems in Earth science because of their devastating consequences. ● Predicting the time remaining before laboratory earthquakes occur from real-time seismic data. ● The data provided was in segments and was messy in nature. ● The signal had a certain time-trend that caused some issues specifically on mean and quantile based features Solution: ● Sampled 10 full earthquakes multiple times (up to 10k times) on Data, and comparing the average KS statistic of all selected features ● Used Matplotlib to visualize the data in every step to extract features. ● We have used LightGBM , Neural networks and XGB Regressor for prediction modeling. Results: ● The overall mean square error score on this was 1.83 ● The solution was awarded by silver medal by kaggle. ● The Solution was in top 2% overall world ranking.
- 14. Challenge: ● Various products of storage systems with various configuration leads to loss track on the health and maintainability ● Customer were unable to take advantage of Software updates due to lack of easy access to information about compatibility ● Failure handling was not precise due to lack of information on the usage trend Solution: Made a common portal for the customers where they can take advantage of different predictive features for all kind of HPE storage systems. Result: ● All kinds of system status are available in the common portal for all of their storage systems which helps in taking various maintenance actions ● Software recommendation helps the users keeping updated their system’s os and different softwares and avoid different anomalies ● Prediction on when the storage might go out of space, when cpu utilization might go at peak or to summarize prediction on when a disaster will happen, helps the users avoid various loses.
- 15. 15 Enable organizations to capture new value and business capabilities Innovation Labs Consistently blogging, to share our knowledge, research Blogs Deeplearning, Coursera, Stanford certified professionals Certifications Insight & perspective to help you to make right business decisions TOK Sessions It’s great to contribute back to the community. We continuously advance open source technologies to meet demanding business requirements. Open Source Contribution
- 16. Machine Learning and AI in Retail
- 21. 21 Eliminating overstocks and out-of-stocks FishEye AI helps retailers replenish supplies by identifying demand for a particular product based on ● sales history ● location ● weather ● promotions ● trends ● … and so on.
- 25. Definition A time series is a set of observation taken at specified times, usually at equal intervals”. “A time series may be defined as a collection of reading belonging to different time periods of some economic or composite variables ● Time series establish relation between “cause” & “Effects”. ● One variable is “Time” which is independent variable & and the second is “Data” which is the dependent variable.
- 26. Examples
- 27. Importance of Time Series Analysis ● Safety from future ● Utility Studies ● Sales Forecasting ● Budgetary Analysis ● Stock Market Analysis ● Stock Market Analysis ● Process and Quality Control ● Inventory Studies ● Economic Forecasting ● Risk Analysis & Evaluation of changes.
- 29. Components of Time Series Is it stationary? Is there a seasonality? Is the target variable autocorrelated?
- 30. Autocorrelation
- 31. Seasonality
- 32. Stationarity
- 34. Modelling time series ModelsARIMA Model Discovery ARCH/ GARCH Model LSTM Autoregres sive or VAR model
- 36. Steps STEP A Data Preprocessing STEP B Data Visualization/ Analysis STEP C Data Inspection STEP D Forecasting results and models A B C D Process Steps Information Knoldus Forecasting platform will take a dataset in any form and load it in its database, and give different option to user for data filtration and preprocessing. In step B, it will give you an Analytics dashboard for reading different aspects from data. After that in Step C , It will give data inspection plot for seasonality, trend and stationarity of data. In final Step D, It will give you all forecasting result and trained models
- 39. Data Inspection
- 40. Conclusion