-
Notifications
You must be signed in to change notification settings - Fork 0
/
index.html
130 lines (119 loc) · 6.97 KB
/
index.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
<!DOCTYPE html>
<!-- saved from url=(0031)https://andrewlaptev.github.io/ -->
<html lang="en" prefix="og: https://ogp.me/ns#"><head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<meta property="og:title" content="NIRMA 621308">
<meta property="og:type" content="website">
<meta property="og:description" content="Project is aimed at stabilizing the psycho-emotional state of users through the personalization of artificial lighting in space">
<meta property="og:url" content="https://andrewlaptev.github.io">
<title>AlgoMath Lab</title>
<style>
body {
max-width: 1000px;
margin: auto;
padding: 50px 50px;
font-family: Arial, Helvetica, sans-serif;
font-size: 19px;
}
a:hover {
color: #57b7eb;
}
a {
color: #4891d3;
}
li {
padding: 5px 0;
}
.header {
display: flex;
flex-direction: column;
align-items: center;
text-align: center;
margin: auto;
}
.logo {
width: 300px;
}
</style>
</head>
<body>
<header class="header">
<img src="logo_pic.jpg" alt="Logo" class="logo">
<h1>AlgoMath Lab of ITMO University</h1>
<h2>Project NIRMA 621305 “Generation
and stratification of data for the evaluation and quality control
of predictive models of the dynamics of socioeconomic systems” (2021-2022)</h2>
</header>
<h3 id="about">About the project</h3>
<p>
The purpose of the project is to develop methods for assessing and controlling
the quality of predictive models (including machine learning models)
of the dynamics of socioeconomic systems.
</p>
<p>
As part of the research work, we develop and test new methods for generating
and stratifying data for assessing and controlling the quality of predictive models
of the dynamics of socioeconomic systems, including models for estimating time of arrival,
link prediction in complex networks; models for predicting the behavior of bank customers
and related financial time series and event chains; recommendation models.
</p>
<h3 id="why">About the methods</h3>
<p>
The methods of data generation and stratification developed during the project are used
to assess and improve the quality of predictive models of evolving complex systems
(transportation in a city, social networks with user profiles, groups of bank customers, etc.),
in particular, they can be used as part of dynamic decision support systems in the areas of
time management, banking, Internet marketing, and chain retail in the presence of newly
incoming data.
</p>
<h3 id="for-whom">Software modules</h3>
<p>The methods are implemented in the form of following modules:</p>
<ul>
<li>
<p>1. The module <a href="https://github.com/AlgoMathITMO/Strat-mETA/">Strat-mETA</a>
for stratification of GPS trajectory data with further estimating time of arrival (ETA)
with a fine-tuned combination of simple regression-based ETA prediction methods.
The results are illustrated on a new real-world car travel dataset of GPS trajectory data.</p>
<p> The companion paper is R. Al Naim, P. Chunaev, K. Bochenina,
<a href="http://urban-computing.com/urbcomp2022/file/UrbComp2022_paper_4809.pdf">«Reproducibility
and Progress in Estimating Time of Arrival, or Can Simple Methods Outperform Deep Learning Ones?»</a>,
The 11th International Workshop on Urban Computing, the 28th ACM SIGKDD 2022 (Best Paper Award Runner Up).
</p>
</li>
<li>
<p>2. The module <a href="https://github.com/AlgoMathITMO/link_prediction/">LinkPredictability</a>
for stratification of predictability (model’s quality level) of links in large non-attributed and node-attributed
networks with further meta-classification. The results are illustrated on synthetic and real-world non-attributed
and node-attributed network datasets.</p>
<p> The companion papers are Stavinova E., Evmenova E., Antonov A., Chunaev P. (2022)
<a href="https://doi.org/10.1007/978-3-030-93409-5_32">«Link Predictability Classes in Complex Networks»</a>,
In: Benito R.M., Cherifi C., Cherifi H., Moro E., Rocha L.M., Sales-Pardo M.
(eds) Complex Networks & Their Applications X. COMPLEX NETWORKS 2021. Studies in Computational Intelligence, vol. 1015. Springer, Cham.
and Antonov, A., Stavinova, E., Evmenova, E., Chunaev., P.
<a href="https://doi.org/10.1007/s13278-022-00912-w">«Link predictability classes in large node-attributed networks»</a>,
Soc. Netw. Anal. Min. 12, 81 (2022).</p>
</li>
<li>
<p>3. The module <a href="https://github.com/AlgoMathITMO/Dynamic-classifier">DynamicTransactionClassifier</a>
for dynamic classification of bank clients by the predictability (model’s quality level)
of their transactional behavior. The results are illustrated on real-world clients’ transactions datasets
in periods with different socioeconomical situation.</p>
<p>The companion paper is Bezbochina, A., Stavinova, E., Kovantsev, A., Chunaev, P. (2022).
<a href="https://doi.org/10.1007/978-3-031-08751-6_36">«Dynamic Classification of Bank Clients by the Predictability of Their Transactional Behavior»</a>,
In: Groen, D., de Mulatier, C., Paszynski, M., Krzhizhanovskaya, V.V., Dongarra, J.J., Sloot, P.M.A.
(eds) Computational Science – ICCS 2022. ICCS 2022. Lecture Notes in Computer Science, vol 13350. Springer, Cham.
</p>
</li>
<li>
<p>4. The module <a href="https://github.com/AlgoMathITMO/SynEvaRec">SynEvaRec</a> for synthetic
data generation applied for evaluating recommender systems in different scenarios.
The results are illustrated on synthetic and real-world user-item-response datasets</p>
<p>The companion paper is V. Provalov, E. Stavinova and P. Chunaev,
<a href="https://doi.org/10.1109/ICDMW53433.2021.00014">«SynEvaRec: A Framework for Evaluating Recommender Systems on Synthetic Data Classes»</a>,
2021 IEEE International Conference on Data Mining Workshops (ICDMW), 2021, pp. 55-64.
</p>
</li>
</ul>
</body></html>