Search

In strategic human resource management, one of the most critical issues to focus on is the correct selection and placement of people. Within the confines of this framework, the reason for the study that was conducted was to explore the machine learning approaches that proved to be the most effective in assisting with the recruitment of personnel and the assessment of their positions. To accomplish this goal, a in a series of tests involving workers in the public sector, categorization algorithms were used. The purpose of these tests was to determine which employees would be the ideal fit in which workstations and to determine how workers should be distributed. For supporting the decision support system, an algorithm model was created. Used in the process of recruiting and evaluating potential workers based on the results of the tests that were given. The most important results of this study support the idea that using the People's Evaluation for Recruitment and Promotion Algorithm Model (EERPAM) would make hiring and promoting people in a company fairer.

The contribution of the paper is the building of a model to predict the proper employees’ allocation in the Greek public sector. To acknowledge the set and weights of criteria upon which our model feeds data, a validated questionnaire was developed and used to conduct a primary quantitative survey amongst HR departments and employees of state organizations in Greece. On the acquired findings, several experiments were administered using linear and machine learning tools, aiming to replace time consuming and subjective procedures, followed by many organizations. Concluding, data classification algorithms are proposed to predict the best matching of employees, giving as inputs personnel qualifications as well as job specifications, leading to a model based on J48, a decision tree algorithm.

Proper selection and positioning of employees is an important issue for strategic human resources management. Within this framework, the aim of the research conducted, was to investigate the most efficient machine learning techniques to support employees’ recruitment and positioning evaluation. Towards this aim, a series of tests were conducted based on classification algorithms concerning employees of the public sector, seeking to predict best fit in workplaces and allocation of employees. Based on the outcome of the administered tests, an algorithm model was built to assist the decision support system of employees’ recruitment and assessment. The primary findings of the present research could lead to the argument that the adoption of the Employees’ Evaluation for Recruitment and Promotion Algorithm Model (EERPAM) will significantly improve the objectivity of employees’ recruitment and positioning procedures.

The Mediterranean Sea is among the most affected areas of our planet by microplastic (MP) pollution. However, some regions are still underrepresented in the current literature. This work studied the fate of microplastics (MPs) released from major populated areas within the NE Ionian Sea, an area that contains highly significant biodiversity. This was accomplished by incorporating oceanographic data into a Lagrangian particle-tracking numerical model that simulated the transport of MP particles for the interval of 27 months. The findings report a high possibility of beaching within the first weeks of the simulation for most locations, where 63 % of MPs were beached and 37 % were still floating at the end of the simulation. Seaward transport and eddy diffusivity are the controlling mechanisms of the MP transport, with diffusion being the primary force controlling the movement of MP particles in 1/3 of the simulated regions. This is highly significant, because in areas where diffusion is the main mechanism controlling MP transport, accumulation of floating MP particles is occurring, as reported in previous studies. The MPs' transport and beaching behavior, as well as the observed residence times, were used to determine the threat level that MPs pose to the biodiversity of specific areas.

Metadata-based similarity measurement is far from obsolete nowadays, despite research's focus on content and context-based information. It allows for aggregating information from textual references, measuring similarity when content is not available, traditional keyword searches in search engines, merging results in meta-search engines, etc. Existing similarity measures do not take into consideration neither the unique nature of multimedia metadata nor the requirements of metadata-based information retrieval of multimedia. This work presents a commonly available author-title multimedia metadata hybrid similarity measure customised that has been shown to be experimentally significantly more effective than baseline measures. In addition, the work presents an architecture and a web-based implementation of an information system for data collection and validation by expert users that allows distributed, binary and scalar result ground-truth definition for a similarity measurement that can be used in digital content's identification and sales.

A preliminary study showed that present performance appraisal systems are considered by the employees, unreliable and subjective and as result they are unable to reveal the real potential of human resources. Employees also consider the existing appraisal procedures bureaucratic, with vague targets and improper evaluation criteria especially formed so as to depict the suitability of employees or future candidates through recruitment procedures to fulfill the positioning needs of a certain job. The present study assessed preliminary data in order: a) to define the parameters and criteria that could improve reliability of evaluation procedures aiming to the selection of the suitable employees or future candidates for certain job positions, and simultaneously b) to clarify which criteria can be measured and must be included in an evaluation system in order to obtain objectivity and meritocracy. A number of such criteria are proposed as a basis of setting the proper procedure of a better, easy to run, transparent, measurable and reliable evaluation system

Olive oil is a key ingredient in the Mediterranean diet and offers many health benefits. However, many factors affect the quality and quantity of olive oil such as olive tree diseases and olive-related pests. Unfortunately, the procedure of identifying pests or the outbreak of a disease is time-consuming, and it depends heavily on the size of the olive grove. Through the use of ICT, remote monitoring of the olive grove can be achieved, by collecting environment-related data and having an overview of the olive grove's overall health. In this paper we propose a low-cost dense network of sensors that collects daily data regarding the olive grove, thus, providing the possibility to prevent infestation of olive fruit fly and/or the outbreak of olive tree-related disease.

Traditionally, the main process for olive fruit fly population monitoring is trap measurements. Although the above procedure is time-consuming, it gives important information about when there is an outbreak of the population and how the insect is spatially distributed in the olive grove. Most studies in the literature are based on the combination of trap and environmental data measurements. Strictly speaking, the dynamics of olive fruit fly population is a complex system affected by a variety of factors. However, the collection of environmental data is costly, and sensor data often require additional processing and cleaning. In order to study the volatility of correlation in trap counts and how it is connected with population outbreaks, a stochastic algorithm, based on a stochastic differential model, is experimentally applied. The results allow us to predict early population outbreaks allowing for more efficient and targeted spraying.

The low cost seismograph under study consists of low cost open source hardware and software microprocessor boards (Arduino Uno R3 and Raspberry Pi 3 B+ ), customized low noise design signal amplifiers, low power dissipation sophisticated power supply, two (2) kinds of earth ground shaking sensors a) Ceramic Accelerometer with cutoff frequency fc=0.15Hz and b) moving coil geophone with cutoff frequency fc=4.5Hz. The signals from the two sensors are amplified independently, while an active second order low-pass anti-alias filter and an 8th order active low-pass anti-alias filter have been used. Finally, a low-cost microprocessor board is responsible for digitizing the analog data from the amplified signal of the sensors with a frequency sampling rate of 345Hz. The aim of the present work is to design and test a systemic protocol in order to evaluate the performance of the proposed low cost seismograph for monitoring local to regional seismicity and micro seismicity. The proposed low-cost system was installed in an area of high seismic activity (Lefkada Island – Village Evgiros) and the recordings are transmitted to the database continuously from the day of its installation up today. Thus have create an amount of data for more than 280 days and all of those data have been stored to our database. Collocated with a high resolution 24 bits digitizer equipped with a broad band seismometer give us the opportunity to compare the recordings. To this end, a testing list of 15 local events has been created with different epicenters and magnitudes. For each event the recording signals have been analyzed in terms of a) power spectrum analysis, b) estimation of first arrival times of both P and S waves, c) signal amplitudes and d) earthquake duration. The choice of those specific measures was done in order to evaluate the performance of the low-cost seismograph in terms of certain seismic parameters such as magnitude, epicenter and source properties. Initial results in terms of the proposed protocol are also presented showing an adequate performance of the propose low cost seismograph. Keywords:low-cost instruments, Ionian islands, performance protocolAcknowledgements«Telemachus – Innovative Seismic Risk Management Operational System of the Ionian Islands» which is part of the Operational Program «Ionian Islands 2014-2020» and is co-financed by the European Regional Development Fund (ERDF) (National Strategic Reference Framework - NSRF 2014-20).

The role of asperities in fault evolution has been received continuously increasing attention as critical areas where nucleation and cascade like failure may take place. They consist patches where the contact takes place across the fault rough surfaces, accumulating elastic strain during the interseismic period. More than one asperity rupture result to strong and large earthquakes, a phenomenon mostly characterizing large subduction earthquakes. Identification of the factors controlling single or multiple asperities failure and their spatiotemporal behaviour is a key issue in seismic hazard assessment. It is the aim of the present work to explore the role of different spatial patters of asperities as well as their different strength characteristics by means of simulation experiments via cellular automata models. Initial results show that the earthquake distribution clearly depends on a) the total real contact area of asperities, b) the relative distance between asperity patches and c) the fraction of strain that asperities may sustain in comparison to the corresponding value of the non–asperity sites. Τhere is a definite range of the aforementioned controling parameters, which result to a non–typical earthquake magnitude distribution and where a clear departure from the classical power law–like Gutenberg – Richter relation is depicted. More specifically, for one (more than one well separated) asperity (–ies) with significant fraction of strain unlocking thresholds a non–typical earthquake size distribution emerges where for low magnitude earthquakes a power law still holds, but for higher earthquake sizes, a quantum like behaviour emerges, i.e. there is one (more than one) certain earthquake sizes that are more probable to occur. This manifests a characteristic earthquake model, which although not adequately supported by observational data, is present in several applications of simulator models. Keywords:asperities, large earthquake quantification, characteristic model, cellular automataAcknowledgements«Telemachus – Innovative Seismic Risk Management Operational System of the Ionian Islands» which is part of the Operational Program «Ionian Islands 2014-2020» and is co-financed by the European Regional Development Fund (ERDF) (National Strategic Reference Framework - NSRF 2014-20).

Olive oil is a key ingredient in the Mediterranean diet and offers many health benefits. However, many factors affect the quality and quantity of olive oil such as olive tree diseases and olive-related pests. Unfortunately, the procedure of identifying pests or the outbreak of a disease is time-consuming, and it depends heavily on the size of the olive grove. Through the use of ICT, remote monitoring of the olive grove can be achieved, by collecting environment-related data and having an overview of the olive grove’s overall health. In this paper we propose a low-cost dense network of sensors that collects daily data regarding the olive grove, thus, providing the possibility to prevent infestation of olive fruit fly and/or the outbreak of olive tree-related disease.

Traditionally, the main process for olive fruit fly population monitoring is trap measurements. Although the above procedure is time-consuming, it gives important information about when there is an outbreak of the population and how the insect is spatially distributed in the olive grove. Most studies in the literature are based on the combination of trap and environmental data measurements. Strictly speaking, the dynamics of olive fruit fly population is a complex system affected by a variety of factors. However, the collection of environmental data is costly, and sensor data often require additional processing and cleaning. In order to study the volatility of correlation in trap counts and how it is connected with population outbreaks, a stochastic algorithm, based on a stochastic differential model, is experimentally applied. The results allow us to predict early population outbreaks allowing for more efficient and targeted spraying.

The aim of this article is to elucidate the identification and investigation of micro-crack evolution as a landslide precursor activity. For this purpose, the construction of a model test was considered appropriate by simulating a soil landslide in a small scale. There is a direct correlation between slope steepness and the occurrence of landslides. When inclination increases, a few seconds before failure, micro-cracks appear, initiating the slide. The whole procedure was recorded by an accelerometer, intending to record micro cracks imprint. The second step upon primary data acquisition was signal analysis in order to locate and examine micro-crack frequency range either a slide occurred and not. Finally, the signal analysis results indicated that there is a specific time period, a few seconds before failure, which, according to its frequency and energy content, can be defined as a landslide precursor activity. Comparing frequency content between precursor activity time period and no activity one greatly can identify the offset difference.