Periodic Reporting for period 1 - Base (Flowbase)
Okres sprawozdawczy: 2019-09-01 do 2020-10-31
Moreover, IT training for employees is not sufficient anymore: People need IT-support directly when they face a problem and are individually customized to their needs. New team members need efficient introduction into organizational structures.
By making software easily accessible for everyone, the benefit for the group or even society is much more likely than when accessibility is limited by complicated UX or UI. If software can only be used after extensive training and practice, it can only be used by people with the sufficient resources, thus making it less accessible for society.
With the technology we have developed at miraminds our clients are able to automatically capture, store, play back, as well as match user software workflows and processes in real-time. Thus developing a technology the replaces the cumbersome manual loading of files from a local file system with an intelligent cloud solution comprising the following offerings/features:
Automated and seamless storing/loading of workflow data on a server/cloud from within our desktop software product.
Intelligent search and/or recommendation of workflows based on a user’s work context.
We consider the course of the project and the work with the Innovation Associate successful. The produced results are in line with what was originally intended and we are thankful for the given opportunity.
However, the transition from a successful research project to an actual product takes time and resources. We have realized that the one-year period of initial research may be too early for developing a sound going-to-market strategy. Due to the successful collaboration, we have invited the Innovation Associate to work for miraminds beyond the funded project period. However, we are now facing the challenge of finding follow up funding for turning gained results into an attractive product.
· Discussing state of miraminds software, requirements and data structure at the time.
· Surveying literature.
· Proposing solutions “on paper” custom to miraminds problem.
· Discussing approaches with miraminds developer.
· Setting up web server on Microsoft Azure.
· Implementing upload web service for data in miraminds flow format (so called reference flows).
· Implementing web service for receiving, storing user tracking data.
· Implementing ‘in-flow’ recommendation web service with a custom nearest neighbor search for matching user tracking data with reference flows in real-tine.
· Implementing web service for storing history of user workflows.
· Implementing web service for ‘next flow’ recommendation based on statistical analysis of user workflow history.
· Implementing dashboard prototype for visualization of obtained data as well as service usage.
· Implementing prototype for the analysis of user workflow divergence from given reference flows.
Work carried out by miraminds developer
· Explaining state of miraminds software, requirements and data structure at the time to IA.
· Discussing proposed approaches with IA.
· Patent research.
· Defining web services and data structures.
· Extending miraminds products with web service connectivity.
· Extending FlowGuide prototype with tracking functionality.
· Extended FlowGuide prototype user interface with visualization of data returned from web services.
Results
With the help of the IA miraminds has developed a machine learning-recommendation system tailored to the data structures and requirements specified by miraminds. The IA implemented his ideas in the form of web services according to interface requirements given by miraminds. miraminds also created test data and client applications to communicate with the IA’s services in order to run realistic tests.
The final milestone of the project (MS5) was a product prototype. By reaching this milestone, miraminds is now able to demonstrate the envisioned recommendation system for software users. A demonstrator/prototype is far from being a viable product. However, with the help of the prototype, miraminds will try to raise funds from stakeholders/companies with an active interest in software user recommendation systems.