KENKO TECHNOLOGY

Key Features

Input

Physiological measurements are collected from different sensors, which can have some form of contact with the user’s body or can be contact free. This data is packed into a chunk of data which includes the user identification, a timestamp indicating when the data was measured, and the data itself.

Input

Algorithm

By receiving the physiological measurements data chunk, our system starts its windowing mechanism and waits for new data to be received. When enough data is available the preprocessing job will take control and pull enough data from the temporary datastore to enrich the Kenko algorithm.

 

Output

each iteration of Kenko Algorithm results with a Kenko Factor that represents the emotional state of the specific user. The Kenko Factor consists of the arousal level and the valence indication. It should be taken into consideration as the system should act to address the user’s emotional load.

Asynchronous

Our API and SDK deal with your data in an asynchronous way, allowing you to add new pieces of data when available, without the application blocking to wait for results. At that time the Kenko Factor is calculated by our unique Kenko algorithm and returned to the system for further actions.

Input

Data management

In order to achieve effective and accurate results, data should be considered as a group of samples and not individual ones. Storing physiological measurements in a temporary data store allows us to rich better analysis based on averages and considering a “bigger picture”. After analysing an individual sample, it is being removed and the data store is now clean of used data.

 

Preprocessing

For better performance and more accurate results, a pre processing phase must be executed on the selected data from the temporary datastore. This preprocessing phase “crunches” the data so our Kenko Algorithm will understand it and use it in the most efficient way possible.

 

Windowing

In order to analyse the data in the most effective manner, Kenko API and SDK uses a sliding time window method. A sliding time window uses time intervals in the data stream to define overlapping bundles of data. For example, a window captures one minute worth of data, but a new window starts every thirty seconds seconds. The frequency with which sliding windows begin is called the period. Therefore, our example would have a window size of one minute and a period of thirty seconds.

 

Because multiple windows overlap, most elements in a data set will belong to more than one window. This kind of Windowing is useful for taking running averages of data.

The diagram can be divided into three different parts:

API/SDK Overview

Product   >

Other emotion analytics methods such as facial expression recognition or speech recognition stores built-in personalized information and has high vulnerability, as shown in Figure 1. Kenko technology utilize physiological data only, which can be stored anonymously and cannot be linked in any way to its source. We do not track the end-users’ personal information and any personal activity in all cases. 

 

Since we do not store any personal data that is related to our end-users, we fully comply with GDPR and other data protection standards and regulations, and promise our customers full protection and confidence in our product.

Figure 1. Comparison of identifiable and unidentifiable biometric data

Kenko API and SDK provides a solution for organizations looking to compose “emotionally aware” applications, customized to internal policies on availability, security, and compliance requirements.

 

With Kenko API and SDK, the “heavy lifting” of understanding the user’s emotions is taken care of, allowing the organization to focus resources on the collection of physiological measurements and performing actions that corresponds to the user emotional situation represented by our algorithm.

 

Using our solution, individuals can become aware of their emotions in real-time and on an ongoing basis. However, each organization has its own requirements and needs. This is why we made Kenko available in two different forms: an online API and an offline SDK.

 

Kenko API: Turn your product “emotionally aware” in minimum effort, by integrating with our online API. It is a quick solution as the integration is easy, straight-forward and does not require any adjustments by both sides.

 

Kenko SDK: Compose “emotionally aware” applications, customized to your organization’s availability, security, and compliance requirements. The SDK is 100% offline and does not require connectivity whatsoever.

 

The diagram below illustrates the functionality available using our product.

API/SDK Overview