Christian Grimme

• Dr.
• Extraordinary Professor of Computational Social Science & Systems Analysis at University of Münster

This work explores the potential to include visual information from images in social media campaign recognition. The diverse content shared on social media platforms, including text, photos, videos, and links, necessitates a multimodal analysis approach. With the emergence of Large Language Models (LLMs), there is now an opportunity to convert imag...

This paper addresses new challenges of detecting campaigns in social media, which emerged with the rise of Large Language Models (LLMs). LLMs particularly challenge algorithms focused on the temporal analysis of topical clusters. Simple similarity measures can no longer capture and map campaigns that were previously broadly similar in content. Here...

We contribute to the efficient approximation of the Pareto-set for the classical NP-hard multiobjective minimum spanning tree problem (moMST) adopting evolutionary computation. More precisely, by building upon preliminary work, we analyze the neighborhood structure of Pareto-optimal spanning trees and design several highly biased sub-graph-based mu...

Social media can be a mirror of human interaction, society, and historic disruptions. Their reach enables the global dissemination of information in the shortest possible time and, thus, the individual participation of people worldwide in global events in almost real-time. However, these platforms can be equally efficiently used in information warf...

We contribute to the efficient approximation of the Pareto-set for the classical $\mathcal{NP}$-hard multi-objective minimum spanning tree problem (moMST) adopting evolutionary computation. More precisely, by building upon preliminary work, we analyse the neighborhood structure of Pareto-optimal spanning trees and design several highly biased sub-g...

The design and choice of benchmark suites are ongoing topics of discussion in the multi-objective optimization community. Some suites provide a good understanding of their Pareto sets and fronts, such as the well-known DTLZ and ZDT problems. However, they lack diversity in their landscape properties and do not provide a mechanism for creating multi...

The rapid spread of disinformation through online environments challenges the development of suitable solution approaches. The scientific evaluation of various intervention strategies shows that until now, no magic bullet has been found that can overcome the problem in all relevant dimensions. Due to the effective impact at the individual level, re...

Traditionally, visualizing benchmark problems is an integral task in the domain of evolutionary algorithms development. Researchers get inspired for new search heuristics by challenges observed in functional landscapes. Moreover, landscape characteristics, features, and even terminology to describe them are derived from visualizations. And most imp...

Today, implications of automation in social media, specifically whether social bots can be used to manipulate people’s thoughts and behaviors are discussed. Some believe that social bots are simple tools that amplify human-created content, while others claim that social bots do not exist at all and that the research surrounding them is a conspiracy...

Single-objective continuous optimization can be challenging, especially when dealing with multimodal problems. This work sheds light on the effects that multi-objective optimization may have in the single-objective space. For this purpose, we examine the inner mechanisms of the recently developed sophisticated local search procedure SOMOGSA. This m...

Recent advances in the visualization of continuous multimodal multi-objective optimization (MMMOO) landscapes brought a new perspective to their search dynamics. Locally eficient (LE) sets, often considered as traps for local search, are rarely isolated in the decision space. Rather, intersections by superposing attraction basins lead to further so...

Recent advances in the visualization of continuous multimodal multi-objective optimization (MMMOO) landscapes brought a new perspective to their search dynamics. Locally efficient (LE) sets, often considered as traps for local search, are rarely isolated in the decision space. Rather, intersections by superposing attraction basins lead to further s...

Multimodality plays a key role as one of the most challenging problem characteristics in the common understanding of solving optimization tasks. Based on insights from the single-objective optimization domain, local optima are considered to be (deceptive) traps for optimization approaches such as gradient descent or different kinds of neighborhood...

This work investigates the functioning of YouTube’s recommendation system with focus on the autoplay function. The autoplay function was often referred to as “radicalizer” in the past, as it was considered to lead towards more extremist content. By an automated data collection through browser remote control, we simulate different usage scenarios (a...

Multi-objective (MO) optimization, i.e., the simultaneous optimization of multiple conflicting objectives, is gaining more and more attention in various research areas, such as evolutionary computation, machine learning (e.g., (hyper-)parameter optimization), or logistics (e.g., vehicle routing). Many works in this domain mention the structural pro...

One of the most significant recent technological developments concerns the development and implementation of ‘intelligent machines’ that draw on recent advances in artificial intelligence (AI) and robotics. However, there are growing tensions between human freedoms and machine controls. This article reports the findings of a workshop that investiga...

Computational social science uses computational and statistical methods in order to evaluate social interaction. The public availability of data sets is thus a necessary precondition for reliable and replicable research. These data allow researchers to benchmark the computational methods they develop, test the generalizability of their findings, an...

Simultaneously visualizing the decision and objective space of continuous multi-objective optimization problems (MOPs) recently provided key contributions in understanding the structure of their landscapes. For the sake of advancing these recent findings, we compiled all state-of-the-art visualization methods in a single R-package (moPLOT). Moreove...

In this work we examine the inner mechanisms of the recently developed sophisticated local search procedure SOMOGSA. This method solves multimodal single-objective continuous optimization problems by first expanding the problem with an additional objective (e.g., a sphere function) to the bi-objective space, and subsequently exploiting local struct...

Simultaneously visualizing the decision and objective space of continuous multi-objective optimization problems (MOPs) recently provided key contributions in understanding the structure of their landscapes. For the sake of advancing these recent findings, we compiled all state-of-the-art visualization methods in a single R-package (moPLOT). Moreove...

Nowadays fake news are heavily discussed in public and political debates. Even though the phenomenon of intended false information is rather old, misinformation reaches a new level with the rise of the internet and participatory platforms. Due to Facebook and Co., purposeful false information - often called fake news - can be easily spread by every...

Multimodality is one of the biggest difficulties for optimization as local optima are often preventing algorithms from making progress. This does not only challenge local strategies that can get stuck. It also hinders meta-heuristics like evolutionary algorithms in convergence to the global optimum. In this paper we present a new concept of gradien...

Visualization techniques for the decision space of continuous multi-objective optimization problems (MOPs) are rather scarce in research. For long, all techniques focused on global optimality and even for the few available landscape visualizations, e.g., cost landscapes, globality is the main criterion. In contrast, the recently proposed gradient f...

The identification of coordinated campaigns within Social Media is a complex task that is often hindered by missing labels and large amounts of data that have to be processed. We propose a new two-phase framework that uses unsupervised stream clustering for detecting suspicious trends over time in a first step. Afterwards, traditional offline analy...

Recently, social bots, (semi-) automatized accounts in social media, gained global attention in the context of public opinion manipulation. Dystopian scenarios like the malicious amplification of topics, the spreading of disinformation, and the manipulation of elections through “opinion machines” created headlines around the globe. As a consequence...

When dealing with continuous single-objective problems, multimodality poses one of the biggest difficulties for global optimization. Local optima are often preventing algorithms from making progress and thus pose a severe threat. In this paper we analyze how single-objective optimization can benefit from multiobjectivization by considering an addit...

Visualization techniques for the decision space of continuous multi-objective optimization problems (MOPs) are rather scarce in research. For long, all techniques focused on global optimality and even for the few available landscape visualizations, e.g., cost landscapes, globality is the main criterion. In contrast, the recently proposed gradient f...

We consider a dynamic bi-objective vehicle routing problem, where a subset of customers ask for service over time. Therein, the distance traveled by a single vehicle and the number of unserved dynamic requests is minimized by a dynamic evolutionary multi-objective algorithm (DEMOA), which operates on discrete time windows (eras). A decision is made...

In practice, e.g. in delivery and service scenarios, Vehicle-Routing-Problems (VRPs) often imply repeated decision making on dynamic customer requests. As in classical VRPs, tours have to be planned short while the number of serviced customers has to be maximized at the same time resulting in a multi-objective problem. Beyond that, however, dynamic...

The detection of orchestrated and potentially manipulative campaigns in social media is far more meaningful than analyzing single account behaviour but also more challenging in terms of pattern recognition, data processing, and computational complexity. While supervised learning methods need an enormous amount of reliable ground truth data to find...

Nowadays fake news are heavily discussed in public and political debates. Even though the phenomenon of intended false information is rather old, misinformation reaches a new level with the rise of the internet and participatory platforms. Due to Facebook and Co., purposeful false information - often called fake news - can be easily spread by every...

Social bots have recently gained attention in the context of public opinion manipulation on social media platforms. While a lot of research effort has been put into the classification and detection of such automated programs, it is still unclear how technically sophisticated those bots are, which platforms they target, and where they originate from...

In online media environments, nostalgia can be used as important ingredient of propaganda strategies, specifically, by creating societal pessimism. This work addresses the automated detection of nostalgic text as a first step towards automatically identifying nostalgia-based manipulation strategies. We compare the performance of standard machine le...

In this work we investigate the engagement of Twitter accounts in the starting phase of reaction cascades, i.e., in the follow-up stream of an original tweet. In a first case study, we focus on a selection of very popular Twitter users from politics and society. We find a small but constantly active set of seemingly automated accounts in the onset...

This book constitutes the refereed proceedings of the First Multidisciplinary International Symposium, MISDOOM 2019, held in Hamburg, Germany, in February/March 2019. The 14 revised full papers were carefully reviewed and selected from 21 submissions. The papers are organized in topical sections named: human computer interaction and disinformation,...

Research has shown that for many single-objective graph problems where optimum solutions are composed of low weight sub-graphs, such as the minimum spanning tree problem (MST), mutation operators favoring low weight edges show superior performance. Intuitively, similar observations should hold for multi-criteria variants of such problems. In this w...

Social bots have recently gained attention in the context of public opinion manipulation on social media platforms. While a lot of research effort has been put into the classification and detection of such (semi-)automated programs, it is still unclear how sophisticated those bots actually are, which platforms they target, and where they originate...

There is a range of phenomena in continuous, global multi-objective optimization, that cannot occur in single-objective optimization. For instance, in some multi-objective optimization problems it is possible to follow continuous paths of gradients of straightforward weighted scalarization functions, starting from locally efficient solutions, in or...

The \(\mathcal {NP}\)-hard multi-criteria shortest path problem (mcSPP) is of utmost practical relevance, e. g., in navigation system design and logistics. We address the problem of approximating the Pareto-front of the mcSPP with sum objectives. We do so by proposing a new mutation operator for multi-objective evolutionary algorithms that solves s...

We continue recent work on the definition of multimodality in multiobjective optimization (MO) and the introduction of a test bed for multimodal MO problems. This goes beyond well-known diversity maintenance approaches but instead focuses on the landscape topology induced by the objective functions. More general multimodal MO problems are considere...

Dieses Kapitel bietet eine Einführung in die Optimierung linearer Zielfunktionen unter Nebenbedingungen. Aufbauend auf einer fundierten Einführung in Notation und grafische Interpretierbarkeit linearer Probleme wird mit der Simplex-Methode nach Dantzig ein effizientes und im Bereich Operations Research zentrales Verfahren zu deren Lösung vorgestell...

Das Kapitel verschafft zuerst einen nicht formalen Einstieg in die Begrifflichkeit der Optimierung und geht dann zur formalen Definition von Optimierungsproblemen über. Die Abstraktion durch Formalität verdeutlicht es an einigen beispielhaften Problemstellungen. Dabei spielen insbesondere auch Themen wie Modellbildung/Abstraktion, Lösungsmethodik u...

Dieses Kapitel widmet sich nichtlinearen Problemstellungen, also jenen Optimierungsproblemen, die entweder in der Problemformulierung selbst oder den Randbedingungen nicht linear sind. Auf eine kurze Rekapitulation von Basiswissen aus der Analysis (Differenziation, Lagrange-Methode) folgt die Vorstellung diverser deterministischer, numerischer Lösu...

Als Abschluss des Buches soll das Kapitel zur Entscheidungstheorie einen anderen, allgemeineren Blickwinkel auf die Problematik der Optimierung vermitteln. Nach einer sehr grundlegenden Einführung in die (traditionelle) Theorie der Entscheidungsfindung sollen die Zusammenhänge zwischen Entscheidungstheorie und Optimierung herausgearbeitet werden. H...

Thema dieses Kapitels sind von der Natur inspirierte algorithmische Konzepte und Verfahren. Aufbauend auf den Ideen der modernen Evolutionstheorie (nach Darwin) und der systematischen Einbindung von Zufall wird in das Gebiet der evolutionären Algorithmen eingeführt. Für diese inzwischen weit verbreiteten und akzeptierten Verfahren wird eine praxiso...

Das Kapitel betrachtet Graphen und Bäume (als spezielle Klasse) zweigeteilt: einerseits als Datenstruktur zur Modellierung von Optimierungsproblemen, auf die ein Optimierungsverfahren angewandt wird (kürzeste Wege, Flussprobleme), andererseits als strukturgebende Elemente für die Konstruktion von Optimierungsverfahren selbst (Strukturierung des Suc...

We analyze the effects of including local search techniques into a multi-objective evolutionary algorithm for solving a bi-objective orienteering problem with a single vehicle while the two conflicting objectives are minimization of travel time and maximization of the number of visited customer locations. Experiments are based on a large set of spe...

The identification of automated activitiy in social media, specifically the detection of social bots, has become one of the major tasks within the field of social media computation. Recently published classification algorithms and frameworks focus on the identification of single bot accounts. Within different Twitter experiments, we show that these...

beschäftigt sich mit verdeckt und automatisiert betriebener Online-Propaganda, Extremismus, Hass und Mobbing in Online-Medien. Dabei stehen einerseits die technischen Möglichkeiten zur Nutzung von Fake-Profilen und Social Bots zur strategischen Manipulation im Fokus. Andererseits werden die zentralen kommunikativen und psychologischen Mechanismen z...

Dieses Lehrbuch vermittelt einen breiten und grundlegenden Einblick in das methodische Verständnis für die Problematik der Optimierung. Im Fokus stehen Algorithmen und Komplexität verschiedener Optimierungsprobleme sowie nützliche Lösungsmethoden und Anwendungsbezug. Dabei wird auf eine ausführliche Darstellung der wichtigen Konzepte der Optimierun...

Social bots are currently regarded an influential but also somewhat mysterious factor in public discourse and opinion making. They are considered to be capable of massively distributing propaganda in social and online media and their application is even suspected to be partly responsible for recent election results. Astonishingly, the term `Social...

Social bots are currently regarded an influential but also somewhat mysterious factor in public discourse and opinion making. They are considered to be capable of massively distributing propaganda in social and online media and their application is even suspected to be partly responsible for recent election results. Astonishingly, the term `Social...

This paper formally defines multimodality in multiobjective optimization (MO). We introduce a test-bed in which multimodal MO problems with known properties can be constructed as well as numerical characteristics of the resulting landscape. Gradient- and local search based strategies are compared on exemplary problems together with specific perform...

The liner shipping fleet repositioning problem (LSFRP) is a central optimization problem within the container shipping industry. Several approaches exist for solving this problem using exact and heuristic techniques, however all of them use a single objective function for determining an optimal solution. We propose a multi-objective approach based...

The acronym EMO stands for a well-established, biennial international conference series devoted to the theory and practice of evolutionary multi-criterion optimization. The first EMO conference took place in 2001 in Zürich (Switzerland), with subsequent conferences taking place in Faro (Portugal), Guanajuato (Mexico), Matsushima (Japan), Nantes (Fr...

The research in evolutionary multi-objective optimization is largely missing a notion of functional landscapes, which could enable a visual understanding of multimodal multi-objective landscapes and their characteristics by connecting decision and objective space. This consequently leads to the negligence of decision space in most algorithmic appro...

Evolutionary algorithms allow for solving a wide range of multi-objective optimization problems. Nevertheless for complex practical problems, including domain knowledge is imperative to achieve good results. In many domains, single-objective expert knowledge is available, but its integration into modern multi-objective evolutionary algorithms (MOEA...

This paper formally defines multimodality in multiobjective optimization (MO). We introduce a test-bed in which multimodal MO problems with known properties can be constructed as well as numerical characteristics of the resulting landscape. Gradient- and local search based strategies are compared on exemplary problems together with specific perform...

One main task in evolutionary multiobjective optimization (EMO) is to obtain a suitable finite size approximation of the Pareto front which is the image of the solution set, termed the Pareto set, of a given multiobjective optimization problem. In the technical literature, the characteristic of the desired approximation is commonly expressed by clo...

We evaluate the performance of a multi-objective evolutionary algorithm on a class of dynamic routing problems with a single vehicle. In particular we focus on relating algo-rithmic performance to the most prominent characteristics of problem instances. The routing problem considers two types of customers: mandatory customers must be visited wherea...

In the a posteriori approach of multiobjective optimization the Pareto front is approximated by a finite set of solutions in the objective space. The quality of the approximation can be measured by different indicators that take into account the approximation's closeness to the Pareto front and its distribution along the Pareto front. In particular...

The algorithmic problem of picking a point uniformly random from a simplex is almost neglected in literature. There are only very few approaches that are limited to a selection from the unit simplex. For an arbitrary simplex no general algorithmic approach is available. In this report, we detail a simple method, which provides the means for selecti...

We propose an evolutionary multiobjective algorithm that approximates multiple reference points (the aspiration set) in a single run using the concept of the averaged Hausdorff distance.

Exploratory Landscape Analysis is an effective and sophisticated approach to characterize the properties of continuous optimization problems. The overall aim is to exploit this knowledge to give recommendations of the individually best suited algorithm for unseen optimization problems. Recent research revealed a high potential of this methodology i...

The incorporation of expert knowledge into multiobjective optimization is an important issue which in this paper is reflected in terms of an aspiration set consisting of multiple reference points. The behaviour of the recently introduced evolutionary multiobjective algorithm AS-EMOA is analysed in detail and comparatively studied for bi-objective o...

In this work, we present an agent-based approach to multi-criteria combinatorial optimization. It allows to flexibly combine elementary heuristics that may be optimal for corresponding single-criterion problems. We optimize an instance of the scheduling problem 1|d j |∑C j ,L max and show that the modular building block architecture of our optimiza...

We present a modular and flexible algorithmic framework to enable a fusion of scheduling theory and evolutionary multi-objective combinatorial optimization. For single-objective scheduling problems, that is the optimization of task assignments to sparse resources over time, a variety of optimal algorithms or heuristic rules are available. However,...

Das Streben nach Verbesserung von Systemen ist schon immer eine allgegenwärtige Herausforderung in Wissenschaft und technischer Anwendung. In der Regel ist jeder Ingenieur nicht nur mit dem Problem konfrontiert, ein einziges Ziel möglichst gut zu erfüllen; für die Entwicklung realer Systeme sind oftmals mehrere sich widersprechende Ziele (oder Krit...

Over the last decade, the predator--prey model (PPM) has emerged as an alternative algorithmic approach to multi-objective evolutionary optimization, featuring a very simple abstraction from natural species interplay and extensive parallelization potential. While substantial research has been done on the former, we for the first time review the PPM...

Standard dominance-based multi-objective evolutionary algorithms hardly allow to integrate problem knowledge without redesigning the approach as a whole. We present a flexible alternative approach based on an abstraction from predator-prey interplay. For parallel machine scheduling problems, we find that the combination of problem knowledge princip...

In production environments, decision makers are often confronted with scheduling problems that demand an optimization of workflow regarding multiple criteria. For specific sub-problems experienced experts have available good heuristics which may contribute to generating a set of multi-criteria solutions. However, current evolutionary multi-criteria...

The Predator Prey Model (PPM) for multi-objective evolutionary optimization features a simple abstraction from natural species interplay: predators represent different objectives and collectively hunt for prey solutions which have to adapt to all predators in order to survive. In this work, we start from previous insights to motivate significant ch...

We present an agent-based approach to multi- objective combinatorial optimization. It allows to flexibly combine elementary heuristics that may be optimal for the corresponding single objective problem. In the multi-objective case a smart com- bination of such heuristics is supposed to efficiently approximate the whole Pareto-front of trade-off sol...

We utilize a competitive coevolutionary algorithm (CA) in order to optimize the parameter set of a Fuzzy System for job negotiation between Community-Grids. In a Community-Grid, users are submitting jobs to their local High Performance Computing (HPC) sites over time. Now, we assume that Community-Grids are interconnected such that the exchange of...

In this paper, we propose a new algorithm for job interchange in Computational Grids that consist of autonomous and equitable HPC sites, called Shaking-G. Originally developed for balancing the sharing of video files in P2P networks, we conceptually transfer and adapt the algorithm to the domain of job scheduling in Grids, building an integrated, l...

In this paper, we address the problem of finding well-performing workload exchange policies for decentralized Computational Grids using an Evolutionary Fuzzy System. To this end, we establish a non-invasive collaboration model on the Grid layer which requires minimal information about the participating High Performance and High Throughput Computing...

In this paper, we address job scheduling in Distributed Computing Infrastructures, that is a loosely coupled network of autonomous acting High Performance Computing systems. In contrast to the common approach of mutual workload exchange, we consider the more intuitive operator’s viewpoint of load-dependent resource reconfiguration. In case of a sit...

In our work, we address the problem of workload distribution within a computational grid. In this scenario, users submit jobs to local high performance computing (HPC) systems which are, in turn, interconnected such that the exchange of jobs to other sites becomes possible. Providers are able to avoid local execution of jobs by offering them to oth...

In our work, we utilize a competitive Co-evolutionary Algorithm in order to optimize the parameter set of a Fuzzy System for job exchange in Computational Grids. In this domain, the providers of High Performance Computing (HPC) centers strive for minimizing the response time for their own customers by trying to distribute workload to other sites in...

In this paper, we address the problem of nding workload exchange policies for decentralized Computational Grids using an Evo- lutionary Fuzzy System. To this end, we establish a non-invasive col- laboration model on the Grid layer which requires minimal information about the participating High Performance and High Throughput Com- puting (HPC/HTC) c...

Traditionally, Predator-Prey Models—although providing a more nature-oriented approach to multi-objective optimization than many other standard Evolutionary Multi-Objective Algorithms—suffer from inherent diversity loss for non-convex problems. Still, the approach to peg single objectives to a predator allows a very simple algorithmic design. The b...

As part of the D-Grid initiative the C3Grid project provides a service oriented Grid infrastructure to support workflow-based scientific applications for Earth System Science. In this paper, we review conceptual as well as technical details of the architecture and implementation concerning this infrastructure and focus especially on the Grid middle...

Implementing efficient data management is a key challenge of grid computing. Due to seemingly different domain specific requirements, data management solutions have been developed separately for each community grid using a selection of low-level tools and APIs. This has led to unnecessarily complex and overspecialized systems.We describe three D-Gr...

In this paper, we apply the parallel predator-prey model for multi-objective optimization to a combinatorial problem for the first time: Exemplarily, we optimize sequences of 50 jobs for an instance of the bi-criteria scheduling problem 1|d j | ∑ C j ,L max with this approach. The modular building block architecture of the predator-prey system and...

In this paper, we introduce a methodology for the approximation of optimal solutions for a resource allocation problem in the domain of Grid scheduling on High Performance Computing systems. In detail, we review a real-world scenario with decentralized, equitable, and autonomously acting suppliers of compute power who wish to collaborate in the pro...

The definition of a generic Grid scheduling architecture is the concern of both the Open Grid Forum’s Grid Scheduling Architecture Research Group and a CoreGRID research group of the same name. Such an architecture should provide a blueprint for Grid system and middleware designers and assist them in linking their scheduling requirements to diverse...