• Company Overview
  • DataBoss is a high-tech research company that provides state-of-the-art data analytics services for cyber security, threat intelligence, big data analysis including model building for regression/classification and stream data processing. The main mission of DataBoss is to provide world class cyber security and data analytics services based on over 20 years of expertise in renowned high-tech companies, Microsoft and IBM, in the USA.

  • We have extensive expertise on:
  • Anomaly Detection
  • Threat Intelligence
  • Forecasting
  • Targeted Ads
  • Machine Learning Algorithms for Commercial Applications
  • Big Data Analytics
  • Text and Semi-structured Data Mining
  • Recommendation Generation
  • Customer Segmentation

Current Projects

Anomaly Based IDS/IPS

Context-aware Anomaly Detection System Based on Big Data Collected from Many Sources

Targeted Online Advertisement Design and Recommendation

Personality Classification for Marketing

Big Data Analytics

Due to outstanding developments in sensor technologies, wide spread usage of smart phones and Internet, we now have the opportunity and capability to gather huge amounts of data in different real life commercial applications (which was not possible in the past). Efficient and effective processing of this big data can significantly improve the performance of many commercial applications. However, this big data has dimensions and volumes unseen before, comes in different modalities and its quality, quantity and statistics rapidly change in time and among elements.

To accommodate these problems, the big data should be processed by advance machine learning algorithms and tools in a streaming manner, i.e., instantly, without any storage requirement, can constantly adapt to the changing statistics or quality of the data, hence can be robust and prone to variations and uncertainties.

We offer full-cycle Data Analytics starting from data gathering and cleansing to profound modeling and suggestions on implementation.

  • We are expert on:
  • Customer Segmentation
  • Knowledge Representation
  • Text and Semi-structured Data Mining
  • Building Statistical Models (e.g., Linear Regression, Logistics Regression, Time Series Models)
  • Targeted Ads
  • Recommendation Generation
  • Building Complex Predictive Models
  • Anomaly Detection
dashboard

Cyber Security

Network Based Intrusion Detection System/Intrusion Prevention System

Your Connection Needs Protection

The amount of cyber attacks has been steadily growing up during the last decade. Network Intrusion Detection and Prevention System (IDS/IPS) can provide a highly effective layer of security designed to protect critical assets from cyber threats.

Databoss provides a completely automated intrusion detection prevention system (IDPS) with real time log analysis for technical and strategic recommendations in order to enhance quality of service. We essentially review your network traffic and data, and can identify probes, attacks, exploits and other vulnerabilities. We warn you of suspicious activities and prevent them. And also, we automatically reconfigure the network to reduce the effects of a suspicious intrusion.

Protect Against Threats with IDS/IPS

Our IDS/IPS Management Solutions provide organizations with full maintenance, updates, rule changes, tuning and 24/7 log monitoring. Clients are able to leverage their current technology investment, using leading security vendors. Our solutions must be properly provisioned, updated and patched to protect against threats. Policies, signatures and rules need to be updated and maintained to ensure accessibility, provide security and to comply with regulations.

  • DataBoss has extensive capabilities in statistical and machine learning based anomaly detection for IPS/IDS with Commercial System Building expertise. DataBoss provides:
  • Network Based IDS based on Deep LSTMs
  • Host Based IDS based on Deep LSTMs
  • Network Behaviour Analysis
  • Log Correlation
  • Threat Intelligence Analysis

HAIKO
Compherensive Automated Smart Personal Assistant

Due to recent advances in device technologies, we now have the ability to collect, store and realtime process considerable amount of data from various sources including cell phone usage statistics to Internet logs. This technical advance is especially noticeable in cell phone technologies. Today, 1/5th of the world’s population use cell phones, out of which 1/3rd are smart phones.


In this product, we build a completely automated smart personal assistance system with real time activity planning and optimization capabilities in order to enhance quality of life. Comprehensive automated smart personal assistance (HAIKO) predicts "the user intend" and act on this intend directly in a time, activity and location aware manner. This predictive capability is essential for a successful automated system with continuous learning capabilities and is achieved by leveraging ideas from machine learning and computational theory.

  • The HAIKO provides:
  • The HAIKO estimates user needs and desires, and it makes life easier by providing related services proactively
  • The HAIKO makes recommendations for restaurants and movies based on the location and desires of the customer
  • The HAIKO provides reminders, search(over the web) events, opportunities and promotions
  • The HAIKO understands through NLP commands, both verbal and written, in the Turkish Language
  • The system has continuous learning abilities in order to enhance its capabilities depending on the changing environment and user needs
  • The HAIKO provides a comprehensive system for predictive transportation optimization
  • The HAIKO schedules meetings
  • The HAIKO performs various tasks:
  • Online search
  • Start and terminate other applications
  • Writing and sending messages
  • Creating alarms, events and tasks
  • Providing weather forecast
  • Making Calls

MULTIVIEW SURVEILLANCE
Context-aware Anomaly Detection System Based on Big Data

Significant developments in information and detection technologies, and increasing use of intelligent mobile devices and internet have bolstered the capacity and capabilities of data acquisition systems beyond expectation. Today, many sources of information from shares on social networks to blogs, from intelligent device activities to security camera recordings are easily accessible.

We implemet state-aware anomaly detection and event prediction systems based on data from diverse sources acquired rapidly under unstationary conditions and quality. The problem of anomaly (or irregularity) detection and subsequent event prediction is a critical topic of national importance, particularly due to increasing border violations and cyber attacks in recent years. However, traditional techniques show insufficient performance in real-life applications due to several reasons:

  • Data acquired from diverse sources are too large in size to be adequately processed by conventional feature extraction, signal processing and machine learning methods.
  • The performance of conventional methods is further impaired by the highly variable properties, structure and quality of text, sound and video data acquired at high speeds. Therefore, highly innovative approaches are urgently needed to cast the full potential of big data on anomaly detection and event prediction applications.


We provide systems that process text on social media, usage statistics of mobile devices and surveillance camera recordings in real time, and subsequently leverage these data for context-dependent anomaly detection and event prediction.

  • Our capabilities include:
  • We scan text in Turkish on social media platforms, and that will extract text features for event detection and prediction.
  • We extract visual features from surveillance camera recordings using deep-learning methods.
  • We process CDR data as well as user information (such as location and time) from mobile devices to perform anomaly detection and event prediction.
  • We perform context-dependent processing and interpretation of data acquired at various dimensions, times and quality from written, audio and visual sources, as well as extracted features.

DEEP ADS
Online Advertisement ForeCasting And Counterfactual Reasoning

In the past 10 years, the number of cell phone users has drastically increased. Thus, delivering the most relevant and up-to-date promotions, information, and advertisements to these users became a significant problem for telecommunication companies since we have a massive amount of relevant data on each user.

In order to obtain a bigger market share, e.g., by providing better advertisement systems, this data should be carefully processed since both user intentions and the characteristics of the data vary from one user to another. Furthermore, the user intentions itself usually change over time, which causes the data to be highly nonstationary. Hence, low complexity sequential machine learning products are needed to intelligently analyze the data, identify the user intentions and trends, predict the most suitable type of advertisement at the most convenient time for each user, and infer the reactions of the users to these advertisements.


We provide a completely automated smart advertising product that choses the most appropriate advertisement for the consumer in the given location, time and event.

  • We model the customers with:
  • Meta data (Age, Gender, Education Status)
  • Mobile Phone (Usage habits, Usage statistics, Social Interaction information)

  • We model the advertisement, new services and the campaigns with these information:
  • Tariff status
  • Price
  • Benefits

Papers

We develop new technologies for cyber security, anomaly detection, big data and data intelligence. Our algorithms are currently used in several different Microsoft and IBM products such as the MSN and the ViaVoice

H. Ozkan, F. Ozkan and S. S. Kozat, "Online Anomaly Detection under Markov Statistics with Controllable Type-I Error," IEEE Transactions on Signal Processing.

Accepted
2015

H. Ozkan, O. Pelvan and S. S. Kozat, "Data Imputation through the Identification of Local Anomalies," IEEE Transactions on Neural Networks and Learning Systems.

Accepted
2015

H. Ozkan, M. A. Donmez, S. Tunc and S. S. Kozat, "A Deterministic Analysis of an Online Convex Mixture of Experts Algorithm," IEEE Transactions on Neural Networks and Learning Systems, vol. 26, no. 7, pp. 1575-1581.

July
2015

N. D. Vanli and S. S. Kozat, "A Unified Approach to Universal Prediction: Generalized Upper and Lower Bounds," IEEE Transactions on Neural Networks and Learning Systems, pp. 646-651.

March
2015

M. O. Sayin, Y. Yilmaz, A. Demir and S. S. Kozat, ``The Krylov-proportionate Normalized Least Mean Fourth Approach: Formulation and Performance Analysis,'' Signal Processing, vol. 109, pp. 1-13.

April
2015

N. D. Vanli, M. A. Donmez and S. S. Kozat, ``Robust Least Squares Methods Under Bounded Data Uncertainties,'' Digital Signal Processing, vol. 36, pp. 82-92.

January
2015

Contact Us

Databoss Security & Analytics
Bilkent Cyberpark
No:4/B Cyberpark Plaza 2. Kat No:B225
Phone: +90 312 290 23 36
Fax: +90 312 290 12 23
Email: info@data-boss.com.tr

DATABOSS is a high-tech research company that provides state-of-the-art data analytics services for cyber security, threat intelligence, data regression/classification, stream data processing. The main mission of DataBoss is to provide world class data analytics services based on over 20 years of expertise in renowned high-tech companies, Microsoft and IBM, in the USA.

© 2016 Databoss Bilisim