On Image Augmentation

The paper considers methods of natural image augmentation, i.e. those whose application results are close to natural impacts on environmental objects that machine learning models may encounter in industrial applications: the influence of weather conditions; operating features or malfunctions of device cameras, etc. The paper presents a taxonomy of methods of natural image augmentation, which includes weather artifacts, camera artifacts, and background substitution for the main object in the image. Existing software libraries for image augmentation are considered in detail, and their shortcomings and limitations are described. The architecture and implementation of a new open library for image augmentation are presented, and the results of its testing on specialized datasets are given. - On Natural Image Augmentation to Increase Robustness of Machine Learning Models

GSMA Open API

GSMA announced the creation of open telephony interfaces for third-party providers.
Lack of third party support has always been the Achilles' heel of telecom, both wired and wireless.
The need for such APIs is obvious, attempts have been made to create them, but there is no result.
Will a new attempt succeed or is it already too late?

from our presentation on FRUCT-2023

Local Services Based on Non-standard Wi-Fi Direct Usage Model

This article discusses a new model for building applied mobile services that use location information and operate in a certain limited spatial area. As a basis for building such applications, a new interpretation of the standard features of Wi-Fi Direct is used. The Wi-Fi Direct specification, in addition to defining the form of device connection, also introduces the concept of a service, when one device offers some service functions to another within the framework of a Wi-Fi Direct connection. Each device can both represent several services and send out search requests for other services. Based on the network proximity architecture, where connections are not used, and wireless network advertising tools are used to convey user information, Wi-Fi Direct services can be considered as key-value databases that exist on mobile devices and can be searched by keys in some local areas. It is these storages that underlie the two models of application services presented in the article that use the spatial proximity of mobile devices: direct messaging between devices without centralized control and the hyper-local Internet model.

source

On Software Defined Radio Issues

Our presentation for MTTW'22

On a formal verification of machine learning systems

The paper deals with the issues of formal verification of machine learning systems. With the growth of the introduction of systems based on machine learning in the so-called critical systems (systems with a very high cost of erroneous decisions and actions), the demand for confirmation of the stability of such systems is growing. How will the built machine learning system perform on data that is different from the set on which it was trained? Is it possible to somehow verify or even prove that the behavior of the system, which was demonstrated on the initial dataset, will always remain so? There are different ways to try to do this. The article provides an overview of existing approaches to formal verification. All the considered approaches already have practical applications, but the main question that remains open is scaling. How applicable are these approaches to modern networks with millions and even billions of parameters? - from our new paper

Analysis and Traffic Management in Smart Cities

This article examines the issues of traffic management in the city. It is well known that cars are a significant contributor to urban air pollution. Accordingly, the management of transport in the city is one of the imperative tasks in terms of supporting the environment and ensuring urban development. Such management is impossible without collecting information on traffic flows in the city. It is logical to assume that there should be a single data source (information store) that contains information about all shipments. Urban governance, of course, involves not only collecting and analyzing data, but also managing this process. This article is devoted to the discussion of such systems. - from our paper

International Conference on Sustainability and Climate Change

City railways – who are their passengers?

In this paper, the approaches to forecasting of passenger flows for new lines of the city railway are analyzed. The article analyzes the experience gained in designing of Moscow central diameters, in the modeling and analysis of their integration with the Moscow Metro and ground urban passenger transport. The main attention is paid to how modern telecommunication technologies make it possible to measure various movements of mobile subscribers, who are passengers, quite accurately. It is the development of telecommunication technologies that is changing the approaches to solving transport problems. What used to be seen as a goal for most transport tasks and was a forecasting goal has now become raw data and can be accurately measured. The article considers what the analysis of transport flows in a modern city now looks like.

On Physical Web for Social Networks

The article discusses the use of Physical Web approaches for the expansion of social networks. This implies the presentation of data from social networks in a real (physical) context, as well as the inverse task of using information about a real physical context in querying and analyzing data from social networks. First of all, mobile phones of social network users are considered as real objects that will be used both in data dissemination and in gathering information about the context. In this case, the purpose of consideration is to build a “natural” extension, when the implementation does not require the creation of a special type of social network entries. The general scheme or model of implementation is based on the minimization (or even complete absence) of requesting additional rights to access the social network, the absence of marks in the social network, and the use of basic functionality and standard protocols for mobile devices.

From our new paper: Namiot D., Sneps-Sneppe M. (2019) On Physical Web for Social Networks. In: Vishnevskiy V., Samouylov K., Kozyrev D. (eds) Distributed Computer and Communication Networks. DCCN 2019. Communications in Computer and Information Science, vol 1141. Springer, Cham

On mobility patterns in Smart City

The article is devoted to the analysis of transport data in the Moscow region, which was used in the design of new urban rail lines. Mobility (smart mobility) is one of the main characteristics of a smart city. Understanding how residents in a city actually move is necessary for the reliable provision of transport services, the construction of new transport lines, etc. Understanding patterns of displacement allows you to identify anomalies in urban displacements, which are evidence of some changes in the urban environment. The basis for the analysis is the data of mobile operators and information on the use of travel documents. - from our new paper

Time series on railways

The article analyzes the patterns of use of railway stations in the Moscow region. The basis for the analysis is the data of smart cards on the entrances and exits of passengers. The technical tool is time series similarity analysis. As a result, the work identifies the main patterns of user behavior on the use of railway stations (railway transport). The results of the work were used in the design of new lines of urban railways. Obviously, the use patterns reflect the current state of the transport system and the urban environment. Accordingly, the recorded changes in usage patterns can serve as indicators and metrics for changes in the urban environment. Our paper in Arxiv

Data Science Books

10 Free Must-Read Books for Machine Learning and Data Science

On transport models

On Passenger Flow Estimation for new Urban Railways

In this paper, we discuss the questions, associated with the forecast for passenger traffic for urban railways. The aim of the study is to select and verify the model for predicting passenger traffic of new urban railways. The article is based on the practical tasks implemented during the project phase for new urban railways in Moscow, Russia. We are considering data sources for building the forecast, as well as practical models that can be used to obtain numerical estimates. Among the discussed data sources, we target migration data that can be collected with the help of telecommunications operators, and information on the use of public transport, obtained from the validation of transport cards. Also, in this paper, we investigate the metrics for traffic along the new city rail line, which can be determined on the basis of the projected passenger traffic. The result of the work was the constructed model of the transport behavior of passengers, taking into account the availability of new urban railways and a set of metrics for assessing the functioning of this transport tool.

More Data Science Books

10 More Free Must-Read Books for Machine Learning and Data Science

5G and Digital Economy

Our presentations from 22st Conference of Open Innovations Association FRUCT:

On 5G projects for urban railways and Russian Digital Economy program

On Digital Economy Issues Looking From the Information Systems Viewpoint

Papers from conrefence: here и here

Procurement assessment

The Art of Service's Procurement Standard Requirements Excel Dashboard and accompanying eBook is for managers, advisors, consultants, specialists, professionals and anyone interested in Procurement assessment. You can see it here

BLE guard

It is a side effect from our experiments with Bluetooth Low Energy tags. See our program BLE guard for Android. You can download it here: http://servletsuite.com/BLECop.apk

The idea of the app is in the continuous scanning (determining of accessibility) for tags and issuing alerts when some changes in the environment are detected. For example, a tag is gone. Or a new tag appears. During the first scan cycle, the app remembers the 'visible’ tags. If during the next consecutive scan any tag is missed, then the application activates the sound alarm and vibration. The typical use cases are:

• You place the tag in your suitcase. If suitcase is “leaving” you – the phone will notify
• Your tag is a bracelet (the collar of the animal). If the bearer of the bracelet moves away from the phone – the phone will notify

And so on.

Also, you can specify the phone number that will receive SMS notification with tag’s loss information. This can be used, for example, for such models: the phone is in the car and the scanner will remember the tags around on the parking slot. If your car gets taken away (evacuated), the phone sends an SMS to a specified number.

By default phone number is blank and application uses only beep and vibration.

Twitter as a transport

Our presentation for AINL-FRUCT conference. You can read the papre here: Namiot D. Twitter as a Transport Layer Platform //arXiv preprint arXiv:1509.06874. – 2015.

Twitter as a Transport Layer Platform from Coldbeans Software

The Physical Web in Smart Cities

Our new paper for RTUWO conference: Dmitry Namiot, Manfred Sneps-Sneppe "The Physical Web in Smart Cities"

In this paper, we discuss the physical web projects based on network proximity for Smart Cities. In general, the Physical Web is an approach for connecting any physical object to the web. With this approach, we can navigate and control physical objects in the world surrounding mobile devices. Alternatively, we can execute services on mobile devices, depending on the surrounding physical objects. Technically, there are different ways to enumerate physical objects. In this paper, we will target the models based on the wireless proximity.

To Deep or not to Deep

An interesting discussion: Will deep learning make other Machine Learning algorithms obsolete?

Our own answer - No. Very often, simpler algorithms like logistic regression will work fine. Deep Learning success depends on the data volume. It should be huge and it is not always true.

Introduction to Markov chain

Markov chain is based on a principle of “memorylessness”. In other words the next state of the process only depends on the previous state and not the sequence of states. This simple assumption makes the calculation of conditional probability easy and enables this algorithm to be applied in number of scenarios. In this article, we will restrict ourself to simple Markov chain. - from here