Jan 23, 2018
Antti Juvonen discusses how to overcome overfitting in deep learning. Check out Antti's data science blog where you can find more articles about deep learning and related topics.
In this post we are going to take a look at a problem called overfitting and it’s potential solution: dropout (or dropout layer). This is relevant for neural networks and deep learning.
First, let’s take a look at the problem that we are trying to get rid of: overfitting. Overfitting happens when the model adapts to training data too well. This usually means that we get excellent results with training data, but poor results with any new data. Overfitting is sometimes called overtraining in machine learni...
Feb 1, 2017
Ymon’s service center starts using innovative technology developed by the Finnish CAP Data Technologies in order to detect abnormal events in information security.
CAP Anomaly Spotter improves detection and response capabilites in order to find potential cybersecurity threats at a service center built by Ymon, a specialist company in networking and information security. The new tool allows Ymon’s situational awareness service to detect hidden abnormalities and to monitor information networks in unprecedented detail.
Normally, maintenance of situational awareness requires creation and management of a multitude of rules to detect the desired events. The new technology allows detection of security events using self-learning artificial intelligence without the need...
Jun 21, 2016
Download the free PDF Brochure and read more about the possibilities of CAP Anomaly Spotter for identity and access management (IAM).
Download the free PDF Brochure and read more about behavioral analytics for remote infrastructure monitoring (RIM).
May 12, 2016
Download the free PDF Brochure and read more about the possibilities of Anomaly Detection for incident response in IT security.
CAP Data Technologies is a Finnish technology startup providing data analytics...
May 12, 2016
Download the free PDF Brochure and read more about the possibilities of Anomaly Detection for Security Operation Centers and MSSPs.
Apr 14, 2016
CAP Anomaly Spotter is now Rsyslog compatible. It is possible to input log lines from Rsyslog sources to CAP Anomaly Spotter and automatically search through vast quantities of log data to identify unusual events in your logs.
Rsyslog is an open source tool that enables logging from many sources, transformations and output to useful destinations.
Contact us for more information about using CAP Anomaly Spotter on-site installation with Rsyslog or other logging platforms.
Tuomo Sipola, CEO
+358 40 753 2169
Nov 24, 2015
CAP Data Technologies has been invited to join Finnish Information Security Cluster, FISC. FISC is an organization established in 2012 by major Finnish information security companies to promote their business and operations in national and international context. Its members are industrial or research organisation providing products and services for various needs in information and cyber security. Most of the member companies operate on international markets.
CAP provides data analytics solutions that reveal unknown unknowns in the data. Our current application sectors include cyber security, financial services and DevOps. The essential element of CAP technology is the ability to detect anomalies in the monitored data without using any pre-set...
Sep 11, 2015
Finnish data analytics startup CAP Data Technologies has secured a €145,000 ($163,000) seed investment round from a group of angel investors. With this boost, the team will continue to develop data analysis services for the global Big Data market.
In this context, business veteran Risto Valtakari complements the board of directors. He brings to CAP Data Technologies his long experience in business development, operations and enterprise IT service sales.
The following business experts have accepted to act as advisors to the management:
Apr 23, 2015
CAP has opened a very simple trial feature where you can get a hands-on touch on what CAP is all about. Click and run, no registration is needed nor disclosing anything about yourself before testing. You may test one of the data-sets we have made available as examples, or you may upload your own data and explore what happens. So, what sort of data can you test?
The primarily use-case is detecting unknown anomalies in server log files. CAP filters out events and lines that do not belong to normal operation. Moreover, you can test CAP with any line-based text file – it may be free text or a numeric excel file saved in .txt format or something that we have not yet figured out ourselves!
CAP is a Big Data analytics service that finds anomalies in the data. Anomaly is de...
Jan 8, 2015
You have a text file with too many lines in it. It would take forever to skim through that file but you need to find strange lines differing from the rest of the file. This article describes how you can detect those anomalies from the file.
For testing purposes you can copy the following lines to a file called access.log.
192.168.0.2 - - [11/May/2014:08:01:19 -0700] "GET /cake.jpg HTTP/1.0" 200 3330 "http://www.example.com/start.html" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_4) AppleWebKit/536.30.1 (KHTML, like Gecko) Version/6.0.5 Safari/536.30.1" "www.example.com" 192.168.0.3 - - [11/May/2014:08:03:01 -0700] "GET /cat.jpg HTTP/1.0" 200 3330 "http://www.example.com/start.html" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_4) AppleWebKit/536.30.1 (KHTML, lik...