Nowaday, emails are used in almost every field, from business to education. Emails have two subcategories, i.e., ham and spam. Email spam, also called junk emails or unwanted emails, is a type of email that can be used to harm any user by wasting his/her time, computing resources, and stealing valuable information. The ratio of spam emails is increasing rapidly day by day. Spam detection and filtration are significant and enormous problems for email and IoT service providers nowadays. Among all the techniques developed for detecting and preventing spam, filtering email is one of the most essential and prominent approaches. Several machine learning and deep learning techniques have been used for this purpose, i.e., Naïve Bayes, decision trees, neural networks, and random forest. This paper surveys the machine learning techniques used for spam filtering techniques used in email and IoT platforms by classifying them into suitable categories. A comprehensive comparison of these techniques is also made based on accuracy, precision, recall, etc. In the end, comprehensive insights and future research directions are also discussed.

/pdf/machine-learning-techniques-for-spam-detection-in-email-and-y5eegmve.pdf

Machine Learning Techniques for Spam Detection in Email and IoT Platforms: Analysis and Research Challenges

Numerous complex systems such as rating systems are highly affected by a small number of spamming attackers. How to design a fast and effective ranking method under the threat of spamming attacks is significant in practice. In this paper, we extract the user’s rating characteristics from personal historical ratings to determine whether the user is normal. It is discovered that reliable users have little bias and their rating scores follow the pattern of peak distribution. On the opposite, malicious users usually have biased ratings and their rating scores scarcely follow a known pattern. A new reputation ranking method IOR (Iterative Optimization Ranking) is proposed based on user rating deviation and rating characteristics. The experimental results on three real datasets show that this method is more efficient than existing states of art methods. This new fundamental idea can be contributed to a new way to solve spammer attacking problem. It can also be applied in large and sparse bipartite rating networks in a short time.

Evaluating user reputation of online rating systems by rating statistical patterns

It is significant to assign reputation scores to users and identify spammers in the bipartite rating networks. In this paper, we propose an Iterative Group-based and Difference Ranking (IGDR) metho...

/pdf/iterative-group-based-and-difference-ranking-method-for-4qmuvkd02k.pdf

Iterative group-based and difference ranking method for online rating systems with spamming attacks

Mammalian DNA folds into 3D structures that facilitate and regulate genetic processes such as transcription, DNA repair, and epigenetics. Several insights derive from chromosome capture methods, such as Hi-C, which allow researchers to construct contact maps depicting 3D interactions among all DNA segment pairs. These maps show a complex cross-scale organization spanning megabase-pair compartments to short-ranged DNA loops. To better understand the organizing principles, several groups analyzed Hi-C data assuming a Russian-doll-like nested hierarchy where DNA regions of similar sizes merge into larger and larger structures. Apart from being a simple and appealing description, this model explains, e.g., the omnipresent chequerboard pattern seen in Hi-C maps, known as A/B compartments, and foreshadows the co-localization of some functionally similar DNA regions. However, while successful, this model is incompatible with the two competing mechanisms that seem to shape a significant part of the chromosomes’ 3D organization: loop extrusion and phase separation. This paper aims to map out the chromosome’s actual folding hierarchy from empirical data. To this end, we take advantage of Hi-C experiments and treat the measured DNA-DNA interactions as a weighted network. From such a network, we extract 3D communities using the generalized Louvain algorithm. This algorithm has a resolution parameter that allows us to scan seamlessly through the community size spectrum, from A/B compartments to topologically associated domains (TADs). By constructing a hierarchical tree connecting these communities, we find that chromosomes are more complex than a perfect hierarchy. Analyzing how communities nest relative to a simple folding model, we found that chromosomes exhibit a significant portion of nested and non-nested community pairs alongside considerable randomness. In addition, by examining nesting and chromatin types, we discovered that nested parts are often associated with active chromatin. These results highlight that crossscale relationships will be essential components in models aiming to reach a deep understanding of the causal mechanisms of chromosome folding.

/pdf/mapping-the-semi-nested-community-structure-of-3d-chromosome-3uvk583o.pdf

Mapping the semi-nested community structure of 3D chromosome contact networks

The operation strategies of a distributed energy system (DES) are usually proposed according to the electrical load (FEL) and the thermal load (FTL), which take the cooling/heating load or electric load as unique constraint conditions that result in a too high or too low equipment load rate. This paper proposes a new hybrid operation strategy (HOS) that takes the full utilization of natural gas and the minimization of power consumption from the power grid as constraints and coordinates the cooling/electricity ratio and heating/electricity ratio of buildings and equipment. In the optimization phase of a DES, an optimization method based on the discretization of the load is proposed to investigate the influence of the uncertainty of the load on the DES, which helps to avoid repeated load simulations and provides stronger adjustability by quoting the normal distribution function to obtain multiple sets of load data with different fluctuations. Further, a multi-objective optimization model combining the genetic algorithm (GA) and mixed integer linear programming algorithm (MILP) was established to find the optimal configuration of equipment capacities by comprehensively considering the annual total cost, carbon emissions, and energy efficiency of the DES. Finally, an office building example was used to validate the feasibility of the above theories and methods. Compared with the FEL and FTL, the HOS reduced the energy waste of the DES by 19.7% and 15.5%, respectively. Compared with using a typical daily load, using an annual hourly load to optimize the DES-HOS produced a better comprehensive performance and lower adverse impacts derived from load fluctuations.

/pdf/optimization-and-performance-analysis-of-a-distributed-1kminph0.pdf

Optimization and Performance Analysis of a Distributed Energy System Considering the Coordination of the Operational Strategy and the Fluctuation of Annual Hourly Load

In the presence of a huge number of possible purchase choices, ranks or ratings of items by others often play very important roles for a buyer to make a final purchase decision. Perfectly objective rating is an impossible task to achieve, and we often use an average rating built on how previous buyers estimated the quality of the product. The problem of using a simple average rating is that it can easily be polluted by careless users whose evaluation of products cannot be trusted, and by malicious spammers who try to bias the rating result on purpose. In this letter we suggest how trustworthiness of individual users can be systematically and quantitatively reflected to build a more reliable rating system. We compute the suitably defined reliability of each user based on the user's rating pattern for all products she evaluated. We call our proposed method as the deviation-based ranking , since the statistical significance of each user's rating pattern with respect to the average rating pattern is the key ingredient. We find that our deviation-based ranking method outperforms existing methods in filtering out careless random evaluators as well as malicious spammers.

https://iopscience.iop.org/article/10.1209/0295-5075/121/68004/pdf

Deviation-based spam-filtering method via stochastic approach

The concept of community detection has long been used as a key device for handling the mesoscale structures in networks. Suitably conducted community detection reveals various embedded informative substructures of network topology. However, regarding the practical usage of community detection, it has always been a tricky problem to assign a reasonable community resolution for networks of interest. Because of the absence of the unanimously accepted criterion, most of the previous studies utilized rather ad hoc heuristics to decide the community resolution. In this work, we harness the concept of consistency in community structures of networks to provide the overall community resolution landscape of networks, which we eventually take to quantify the reliability of detected communities for a given resolution parameter. More precisely, we exploit the ambiguity in the results of stochastic detection algorithms and suggest a method that denotes the relative validity of community structures in regard to their stability of global and local inconsistency measures using multiple detection processes. Applying our framework to synthetic and real networks, we confirm that it effectively displays insightful fundamental aspects of community structures.

Consistency landscape of network communities.

We use the structure of a real distribution network of agricultural product in Korea and investigate how the change in the supply may affect the price changes in agents across the distribution network. In particular, we focus on the real network structure of cabbage distribution composed of various types of agents, from farms to consumers, and apply a dynamic model to describe how each participant reacts upon the change of input and output flow of products through the adjustment of price. Our main result implies that the effect of fluctuation of production quantity in the supplying participant can be nontrivial and the consumer price responds to such changes. We believe that our results can be useful to predict what will happen if the agricultural production changes much in the future due to the climate changes.

Product flow and price change in an agricultural distribution network

We propose an extended genetic algorithm (GA) with different local environmental conditions. Genetic entities, or configurations, are put on nodes in a ring structure, and location-dependent environmental conditions are applied for each entity. Our GA is motivated by the geographic aspect of natural evolution: Geographic isolation reduces the diversity in a local group, but at the same time, can enhance intergroup diversity. Mating of genetic entities across different environments can make it possible to search for broad area of the fitness landscape. We validate our extended GA for finding the ground state of three-dimensional spin-glass system and find that the use of different environmental conditions makes it possible to find the lower-energy spin configurations at relatively shorter computation time. Our extension of GA belongs to a meta-optimization method and thus can be applied for a broad research area in which finding of the optimal state in a shorter computation time is the key problem.

Daekyung Lee

Papers

Deviation-based spam-filtering method via stochastic approach

Consistency landscape of network communities.

Product flow and price change in an agricultural distribution network

Different Environmental Conditions in Genetic Algorithm