Dmitry Zinoviev | Suffolk University Boston MA (original) (raw)

Uploads

Videos by Dmitry Zinoviev

My course "Introduction to Complex Networks for Programmers and Non-Programmers" IN RUSSIAN is of... more My course "Introduction to Complex Networks for Programmers and Non-Programmers" IN RUSSIAN is officially available on Udemy (https://www.udemy.com/course/dzcnapy_ru/learn/lecture/26242502) The course covers #python (#networkx and #pandas) and #gephi.

2 views

Books by Dmitry Zinoviev

Make your good Python code even better by following proven and effective pythonic programming tip... more Make your good Python code even better by following proven and effective pythonic programming tips. Avoid logical errors that usually go undetected by Python linters and code formatters, such as frequent data look-ups in long lists, improper use of local and global variables, and mishandled user input. Discover rare language features, like rational numbers, set comprehensions, counters, and pickling, that may boost your productivity. Discover how to apply general programming patterns, including caching, in your Python code. Become a better-than-average Python programmer, and develop self-documented, maintainable, easy-to-understand programs that are fast to run and hard to break.

Reusing well-written, well-debugged, and well-tested code improves productivity, code quality, an... more Reusing well-written, well-debugged, and well-tested code improves productivity, code quality, and software configurability and relieves pressure on software developers. When you organize your code into self-contained modular units, you can use them as building blocks for your future projects and share them with other programmers, if needed. Understand the benefits and downsides of seven code reuse models so you can confidently reuse code at any development stage. Create static and dynamic libraries in C and Python, two of the most popular modern programming languages. Adapt your code for the real world: deploy shared functions remotely and build software that accesses them using remote procedure calls.

The Routledge Handbook of Service Research Insights and Ideas, 2020

The topic of service innovation (SI) attracts substantial interest among both service researchers... more The topic of service innovation (SI) attracts substantial interest among both service researchers and industrial practitioners. Drawing from network theory and complex network analysis methods, this chapter maps the conceptual structure of SI by detecting and delineating related themes and their interconnections. Based on subject terms collected from more than two thousand SI-related articles published in 786 scholarly journals during the years from 2003 to 2018, six SI clusters or domains are identified. Next, the critical concepts, typical research methods, and prominent industry contexts of each domain are identified and used to inform a multilevel SI conceptual framework. This chapter closes with a discussion on the proposed structural holes that imply novel, combinative research opportunities between otherwise disparate domains for future SI research.

Go from messy, unstructured artifacts stored in SQL and NoSQL databases to a neat, well-organized... more Go from messy, unstructured artifacts stored in SQL and NoSQL databases to a neat, well-organized dataset with this quick reference for the busy data scientist. Understand text mining, machine learning, and network analysis; process numeric data with the NumPy and Pandas modules; describe and analyze data using statistical and network-theoretical methods; and see actual examples of data analysis at work. This one-stop solution covers the essential data science you need in Python.

The issue of information diffusion in small-world social networks was first systematically brough... more The issue of information diffusion in small-world social networks was first systematically brought to light by Mark Granovetter in his seminal paper “The Strength of Weak Ties” in 1973 and has been an area of active academic studies in the past three decades. In this review, I discuss information proliferation mechanisms in massive online social networks (MOSNs—networks with thousands and millions of members). In particular, I address the following aspects of information diffusion processes: the role and the strategic position of influential spreaders of information; the pathways in the social networks that serve as conduits for communication and the flow of information; mathematical models describing the proliferation processes; short-term and longterm dynamics of information diffusion, and secrecy of information diffusion.

Papers by Dmitry Zinoviev

ArXiV Preprint, 2023

Elon Musk has long been known to significantly impact Wall Street through his controversial state... more Elon Musk has long been known to significantly impact Wall Street through his controversial statements and actions, particularly through his own use of social media. An innovator and visionary entrepreneur, Musk is often considered a poster boy for all entrepreneurs worldwide. It is, thus, interesting to examine the effect that Musk might have on Main Street, i.e., on the social media activity of other entrepreneurs. In this research, we study and quantify this "Musk Effect," i.e., the impact of Musk's recent and highly publicized acquisition of Twitter on the tweeting activity of entrepreneurs. Using a dataset consisting of 9.94 million actual tweets from 47,190 self-declared entrepreneurs from seven English-speaking countries (US,

A COOL-1 multiprocessor shared memory system based on superconductor Rapid Single-Flux Quantum (R... more A COOL-1 multiprocessor shared memory system based on superconductor Rapid Single-Flux Quantum (RSFQ) technology is being developed at SUNY (Stony Brook, USA) within the framework of the Hybrid Technology Multithreaded architecture (HTMT) petaflops project led by JPL. This paper describes a multithreading approach proposed in the COOL-I architecture and mechanisms to exploit the thread level parallelism in RSFQ processors called SPELL-1. Up to 128 fine-grain threads called (instruction) streams arranged in 16 groups of 8 streams each can run in parallel within a SPELL-1 processor. All eight streams comprising each COOL stream cluster can communicate and synchronize directly via shared registers. Fast creation and termination of streams including speculative stream execution are also supported.

Design and partial implementation of RSFQ-based Batcher-Banyan switch and support tools. Dmitry Y... more Design and partial implementation of RSFQ-based Batcher-Banyan switch and support tools. Dmitry Yurievich Zinoviev State University of New York at Stony Brook, 1997.

Page 1. RSFQ Pseudo Random Generator and Its Possible Applications Anna Yu. Kidiyarova-Shevchenko... more Page 1. RSFQ Pseudo Random Generator and Its Possible Applications Anna Yu. Kidiyarova-Shevchenko Institute of Nuclear Physics, Moscow State University, Vorobiyovy Gory, Moscow, Russia Dmitry Yu. Zinoviev Department ...

In this paper, I present the design and implementation of Clown--a simulator of a microprocessor-... more In this paper, I present the design and implementation of Clown--a simulator of a microprocessor-based computer system specifically optimized for teaching operating system courses at undergraduate or graduate levels. The package includes the simulator itself, as well as a collection of basic I/O devices, an assembler, a linker, and a disk formatter. The simulator architecturally resembles mainstream microprocessors from the Intel 80386 family, but is much easier to learn and program. The simulator is fast enough to be used as an emulator--in the direct user interaction mode.

This paper describes the results of a preliminaryanalysis of ultra-fast low-power superconductord... more This paper describes the results of a preliminaryanalysis of ultra-fast low-power superconductordigital switches based on Rapid Single-Flux-Quantum(RSFQ) technology. In particular, RSFQ-based crossbar,Batcher-banyan, and shared bus switching cores havebeen considered, and the possible parameters of these circuitshave been estimated. The results show that theproposed RSFQ digital switches with overall throughputof 7:5 T bps operating at an internal clock frequency of? 60

2 views

The Routledge Handbook of Service Research Insights and Ideas, 2020

ArXiV Preprint, 2023

Proc Spie, Feb 19, 1999

We have carried out a paper feasibility study of the implementation of most common packet switchi... more We have carried out a paper feasibility study of the implementation of most common packet switching cores (crossbar, Batcher-banyan, time-division shared bus, and token ring) using the superconductor rapid single flux quantum (RSFQ) digital technology. According to our estimates, the best performance-to-complexity ratio may be obtained for the Batcher-banyan network. For example, a 128 by 128 switching core with self-routing

Eprint Arxiv Cs 0508128, Aug 30, 2005

Persons who engage in non-suicidal self-injury (NSSI), often conceal their practices which limits... more Persons who engage in non-suicidal self-injury (NSSI), often conceal their practices which limits the examination and understanding of those who engage in NSSI. The goal of this research is to utilize public online social networks (namely, in LiveJournal, a major blogging network) to observe the NSSI population's communication in a naturally occurring setting. Specifically, LiveJournal users can publicly declare their interests. We collected the self-declared interests of 22,000 users who are members of or participate in 43 NSSI-related communities. We extracted a bimodal socio-semantic network of users and interests based on their similarity. The semantic subnetwork of interests contains NSSI terms (such as "self-injury" and "razors"), references to music performers (such as "Nine Inch Nails"), and general daily life and creativity related terms (such as "poetry" and "boys"). Assuming users are genuine in their declarations, the words reveal distinct patterns of interest and may signal keys to NSSI.

Instant quality feedback in the form of online peer ratings is a prominent feature of modern mass... more Instant quality feedback in the form of online peer ratings is a prominent feature of modern massive online social networks (MOSNs). The feedback is typically implemented as a Facebook-style “Like” mechanism that allows users to indicate their appreciation of a post, comment, photograph, etc. We found that bipolar online peer rating systems can be efficiently used to deliver user feedback.

4th IC2S2, 2018

Donald Trump has been an avid user of Twitter before, throughout, and in the aftermath of the 201... more Donald Trump has been an avid user of Twitter before, throughout, and in the aftermath of the 2017 USA presidential election campaign. Secretary Hillary Clinton, the Democratic Party candidate, was active on Twitter only from the beginning to end of the campaign. The goal of this research is to reconstruct the timeline and logic of the campaign using complex network analysis of President Trump's and Secretary Clinton's tweets.

North Korea (DPRK) escalates its nuclear program. China is believed to have influence over DPRK. ... more North Korea (DPRK) escalates its nuclear program. China is believed to have influence over DPRK. China tightens the screws on DPRK. DPRK does not seem to care. Does China really have influence over DPRK?

The concept of fraud, its antecedents and outcomes as well as its detection and prevention, have ... more The concept of fraud, its antecedents and outcomes as well as its detection and prevention, have been discussed by both academics and practitioners for decades. The scope and complexity of the concept of fraud attracts scholars from diverse disciplines.

The purpose of our study is to gain a broader understanding of how fraud is viewed in the compendium of academic literature. Using semantic network analysis method, we explore the structure of fraud-related research and analyze the internal connections among the current areas of interest for fraud researchers. We are particularly interested in the relationship between the mainstream financial accounting and audit research and the field of information systems and technology.

Our work makes a big stride toward the understanding of current state of fraud-related research. The interdisciplinary semantic map of keywords and subject terms helps understand the trends in fraud scholarship, identify gaps and propose directions for future research.

Conspiracy theories (including those related to secret societies) and pseudo-sciences (CT&PS) are... more Conspiracy theories (including those related to secret societies) and pseudo-sciences (CT&PS) are as old as the humankind. The abundance of CT&PS surviving in the modern world is astonishing: Wikipedia [Wik16a] mentions 200 categories of conspiracy theories and secret societies, and ca. 150 topics classified as pseudoscience [Wik16b]. The goal of our project is to quantify relationships among these topics, on one hand, and between these topics and other, non-conspiratorial and non-pseudo-scientific, topics, on the other hand. We use title and co-purchasing information from Amazon.com—a new source of cheap and relatively easily collectible research data [RW14]. The dataset consists of 101,000 distinct titles, mostly referring to books and collected in August–December 2016 by starting with a manually selected collection of seed titles and following the links to 4–10 titles that are frequently purchased together. About 9,500 titles in the dataset are related to CT&PS. We manually assigned each CT&PS title to at least one of the 360 categories from the Wikipedia lists [Wik16a, Wik16b]. (We plan to use these tagged titles in the future in machine learning-based classification projects.) We arranged the categories into a network, where two categories are connected if the titles in these categories are frequently purchased together (as reported by Amazon.com). Since Amazon.com shows only the first few related items, our network is sparser than the real latent co-purchase network. The resultant network consists of 267 categories and 99 connected components. 93 categories were not included because there had few or no titles in them. Most components have 4 or fewer categories (e.g., " zang fu–meridian therapy–cupping therapy–moxibustion " or " big government "), but the largest connected component—the giant component—has 152 (57%) categories. We hypothesize that the small isolated components represent distinct CT&PS topics—bags of categories that describe purchase baskets of customers interested in the field. However, the same reasoning cannot be extended to the giant component. We used louvain network community detection algorithm [BGLL08] to partition the giant component into 15 modular communities—sets of nodes that have more connections among them than to the other sets. We hypothesize that each community is yet another form of a distinct CT&PS topic.

More than 4,600 non-academic music groups emerged in the USSR and post-Soviet independent nations... more More than 4,600 non-academic music groups emerged in the USSR and post-Soviet independent nations in 1960–2015, performing in 275 genres and sub-genres, including rock, pop, disco, jazz, and folk. Some of the groups became legends and survived for decades, while others vanished and are known now only to select music history scholars and fans. The total number of unique performers in all groups exceeds 17,000, and at least 3,600 of them participated in more than one project. The goal of this study is to investigate if sharing performers with other groups influences the groups' eventual success, and try to predict the success, based on performers sharing. We collected information about 4,560 Russian, Ukrainian, Belorussian, Estonian, Latvian, Lithuanian, and Moldavian non-academic music groups that have one or more Wikipedia pages in any language. We built a network of the groups by representing groups as network nodes and connecting two groups with an arc if they shared at least one performer. The total number of recorded shares is 16,329. For each node in the network, we calculated degree, average neighbor degree, closeness, betweenness, and eigenvalue centrality, and the clustering coefficient. These numbers represent various aspects of performers' sharing, such as openness to sharing (degree), prestige (eigenvalue), and embeddedness (clustering coefficient). In the absence of sales and chart data caused in the first place by the lack of proper music entertainment industry in the USSR and informal status of most of the groups in the Soviet era, we propose to use secondary but easily collectible descriptors as proxies to the groups' success. These descriptors are the combined group Wikipedia pages visit frequencies (in all languages in the last three years) and maximum Google page rank. We selected approximately 2,000 groups for further analysis. We discovered that both Google page rank and visit frequency are positively (but insignificantly) correlated with all four centrality measures and the average neighbor degree, and negatively (but insignificantly) correlated with the clustering coefficient. We also discovered that the six network measures together serve as reasonably accurate predictors of the page rank and visit frequency. The accuracy of our random forest-based predictive model is 71% for the visit frequency range (below median vs above median) and 49% for the Google page rank (94% if an error of ±1 is allowed). We hypothesize that music groups benefit from the cultural cross-pollination caused by performers moving between different projects. However, these transfers have a limited scope. We calculated the genre-based generalized similarity for every pair of connected groups (sharing at least one musician) and unconnected groups and discovered that the connected groups on average perform in more similar genres than unconnected groups. Likewise, groups performing in similar genres are slightly more likely to share musicians. We believe that our genre classification mechanism, based on generalized similarity, and the network-based success explanation (and possibly prediction) methods are easily extensible to other areas of arts and humanities that have genre-oriented structure and medium-to long-term team-based collaborations.

"Kompromat" (the Russian word for "compromising material") has been efficiently used to harass Ru... more "Kompromat" (the Russian word for "compromising material") has been efficiently used to harass Russian political and business elites since the days of the USSR. Online crowdsourcing projects such as "RuCompromat" made it possible to catalog and analyze kompromat using quantitative techniques-namely, social network analysis. In this paper, we constructed a social network of 11,000 Russian and foreign nationals affected by kompromat in Russia in 1991-2020. The network has an excellent modular structure with 62 dense communities. One community contains prominent American officials, politicians, and entrepreneurs (including President Donald Trump) and appears to concern Russia's controversial interference in the 2016 U.S. presidential elections. Various network centrality measures identify seventeen most central kompromat figures, with President Vladimir Putin solidly at the top. We further reveal four types of communities dominated by entrepreneurs, politicians, bankers, and law enforcement officials ("siloviks"), the latter disjointed from the first three.

—More than 4,600 non-academic music groups emerged in the USSR and post-Soviet independent nation... more —More than 4,600 non-academic music groups emerged in the USSR and post-Soviet independent nations in 1960–2015, performing in 275 genres. Some of the groups became legends and survived for decades, while others vanished and are known now only to select music history scholars. We built a network of the groups based on sharing at least one performer. We discovered that major network measures serve as reasonably accurate predictors of the groups' success. The proposed network-based success exploration and prediction methods are transferable to other areas of arts and humanities that have medium-or long-term team-based collaborations.

Survivalists, or preppers, are individuals or groups who are actively preparing for emergencies, ... more Survivalists, or preppers, are individuals or groups who are actively preparing for emergencies, including possible disruptions in social or political order, on scales from local to international. (In their terminology, for when " shit hits the fan, " SHTF.) Picture 1 shows a network of prepper-related topics based on 2,000 titles sold on Amazon.com and collected in 2016–17, that either deal with preppers' issues or are purchased together with other preppers' titles.

The paper shows a method of forming a table of 34 qualitative adjectives of the Russian language.... more The paper shows a method of forming a table of 34 qualitative adjectives of the Russian language. This table can be used to create ordinal scales with ten or less levels; for example, a ten-stop scale "excellent - excellent - strong - solid - good - so so - weak - bad - awful - disgusting."