Search results for: decision tree classifier

Authors: Noppadol Sritragool

Abstract:

The objectives of this research were to investigate the decision of passengers who choose to fry with low cost airlines and to study marketing factors which have the influence to the decision to choose each low cost airlines. This paper was a quantitative research technique. A total of 400 low cost airlines’ passengers were interviewed via English questionnaire to collect the respondents’ opinions. The findings revealed that respondents were male and female at a similar proportion. The majority had at least an undergraduate degree, have a lower management level jobs, and had income in the range of 25,000 -35,000 baht per month.. In addition, the findings also revealed that the first three marketing factors influencing the decision of the respondents to choose low-cost airlines were low price, direct flight, and online system.

Keywords: decision to choose, marketing factors, low-cost airlines

Procedia PDF Downloads 428

Authors: E. Younsi, H. Andriamboavonjy, A. David, S. Dokou, B. Lemrabet

Abstract:

Software and time optimization are very important factors in financial markets, which are competitive fields, and emergence of new computer tools further stresses the challenge. In this context, any improvement of technical indicators which generate a buy or sell signal is a major issue. Thus, many tools have been created to make them more effective. This worry about efficiency has been leading in present paper to seek best (and most innovative) way giving largest improvement in these indicators. The approach consists in attaching a signature to frequent market configurations by application of frequent patterns extraction method which is here most appropriate to optimize investment strategies. The goal of proposed trading algorithm is to find most accurate signatures using back testing procedure applied to technical indicators for improving their performance. The problem is then to determine the signatures which, combined with an indicator, outperform this indicator alone. To do this, the FP-Tree algorithm has been preferred, as it appears to be the most efficient algorithm to perform this task.

Keywords: quantitative analysis, back-testing, computational models, apriori algorithm, pattern recognition, data mining, FP-tree

Procedia PDF Downloads 362

Authors: Szymon Rutkowski, Tomasz Korbak

Abstract:

We present a reinforcement-learning-based method of training neural machine translation models without parallel corpora. The standard encoder-decoder approach to machine translation suffers from two problems we aim to address. First, it needs parallel corpora, which are scarce, especially for low-resource languages. Second, it lacks psychological plausibility of learning procedure: learning a foreign language is about learning to communicate useful information, not merely learning to transduce from one language’s 'encoding' to another. We instead pose the problem of learning to translate as learning a policy in a communication game between two agents: the translator and the classifier. The classifier is trained beforehand on a natural language inference task (determining the entailment relation between a premise and a hypothesis) in the target language. The translator produces a sequence of actions that correspond to generating translations of both the hypothesis and premise, which are then passed to the classifier. The translator is rewarded for classifier’s performance on determining entailment between sentences translated by the translator to disciple’s native language. Translator’s performance thus reflects its ability to communicate useful information to the classifier. In effect, we train a machine translation model without the need for parallel corpora altogether. While similar reinforcement learning formulations for zero-shot translation were proposed before, there is a number of improvements we introduce. While prior research aimed at grounding the translation task in the physical world by evaluating agents on an image captioning task, we found that using a linguistic task is more sample-efficient. Natural language inference (also known as recognizing textual entailment) captures semantic properties of sentence pairs that are poorly correlated with semantic similarity, thus enforcing basic understanding of the role played by compositionality. It has been shown that models trained recognizing textual entailment produce high-quality general-purpose sentence embeddings transferrable to other tasks. We use stanford natural language inference (SNLI) dataset as well as its analogous datasets for French (XNLI) and Polish (CDSCorpus). Textual entailment corpora can be obtained relatively easily for any language, which makes our approach more extensible to low-resource languages than traditional approaches based on parallel corpora. We evaluated a number of reinforcement learning algorithms (including policy gradients and actor-critic) to solve the problem of translator’s policy optimization and found that our attempts yield some promising improvements over previous approaches to reinforcement-learning based zero-shot machine translation.

Keywords: agent-based language learning, low-resource translation, natural language inference, neural machine translation, reinforcement learning

Procedia PDF Downloads 128

Authors: Diego De Castro Rodrigues, Marcelo B. Nery, Sergio Adorno

Abstract:

The project aims to develop a framework for social data analysis, particularly by mobilizing criminal records and applying descriptive computational techniques, such as associative algorithms and extraction of tree decision rules, among others. The methods and instruments discussed in this work will enable the discovery of patterns, providing a guided means to identify similarities between recurring situations in the social sphere using descriptive techniques and data visualization. The study area has been defined as the city of São Paulo, with the structuring of social data as the central idea, with a particular focus on the quality of the information. Given this, a set of tools will be validated, including the use of a database and tools for visualizing the results. Among the main deliverables related to products and the development of articles are the discoveries made during the research phase. The effectiveness and utility of the results will depend on studies involving real data, validated both by domain experts and by identifying and comparing the patterns found in this study with other phenomena described in the literature. The intention is to contribute to evidence-based understanding and decision-making in the social field.

Keywords: social data analysis, criminal records, computational techniques, data mining, big data

Procedia PDF Downloads 85

Authors: Djamila Meghraoui, Bachir Boudraa, Thouraya Meksen, M.Boudraa

Abstract:

Parkinson's disease is a pathology that involves characteristic perturbations in patients’ voices. This paper describes a proposed method that aims to diagnose persons with Parkinson (PWP) by analyzing on line their voices signals. First, Thresholds signals alterations are determined by the Multi-Dimensional Voice Program (MDVP). Principal Analysis (PCA) is exploited to select the main voice principal componentsthat are significantly affected in a patient. The decision phase is realized by a Mul-tinomial Bayes (MNB) Classifier that categorizes an analyzed voice in one of the two resulting classes: healthy or PWP. The prediction accuracy achieved reaching 98.8% is very promising.

Keywords: Parkinson’s disease recognition, PCA, MDVP, multinomial Naive Bayes

Procedia PDF Downloads 278

Authors: Benlakhdar Mohamed Larbi, Yagoub Asma

Abstract:

Purpose: This study aims at highlighting the significance and importance of utilizing marketing information system (MKIS) on decision-making, by clarifying the need for quick and efficient decision-making due to time saving and preventing of duplication of work. Design, methodology, approach: The study shows the roles of each part of MKIS for developing marketing strategy, which present a real challenge to individuals and institutions in an era characterized by uncertainty and clarifying the importance of each part separately, depending on decision type and the nature of the situation. The empirical research method was evaluated by specialized experts, conducted by means of questionnaires. Correlation analysis was employed to test the validity of the procedure. Results: The empirical study findings confirmed positive relationships between the level of utilizing and adopting ‘decision support system and marketing intelligence’ and the success of an organizational decision-making, and provide the organization with a competitive advantage as it allows the organization to solve problems. Originality/value: The study offer better understanding of performance- increasing market share as an organizational decision making based on marketing information system.

Keywords: database, marketing research, marketing intelligence, decision support system, decision-making

Procedia PDF Downloads 330

Authors: Rolla Monib, Chris Ian Goodier, Alistair Gibb

Abstract:

This paper aims to investigate decision support strategies in the EC sector to determine the most appropriate degree of modularization. This is achieved through three oil and gas (O&G) and two power plant case studies via semi-structured interviews (n=59 and n=27, respectively), analysis of project documents, and case study-specific semi-structured validation interviews (n=12 and n=8). New terminology to distinguish degrees of modularization is proposed, along with a decision-making support checklist and a diagrammatic decision-making support figure. Results indicate that the EC sub-sectors were substantially more satisfied with the application of component, structural, or traditional modularization compared with system modularization for some types of modules. Key drivers for decisions on the degree of modularization vary across module types. This paper can help the EC sector determine the most suitable degree of modularization via a decision-making support strategy.

Keywords: modularization, engineering construction, case study, decision support

Procedia PDF Downloads 94

Authors: Xiuying Wang, Changliang Li, Bo Xu

Abstract:

A fixed-length feature vector is required for many machine learning algorithms in NLP field. Word embeddings have been very successful at learning lexical information. However, they cannot capture the compositional meaning of sentences, which prevents them from a deeper understanding of language. In this paper, we introduce a novel hierarchical tree long short-term memory (HTLSTM) model that learns vector representations for sentences of arbitrary syntactic type and length. We propose to split one sentence into three hierarchies: short phrase, long phrase and full sentence level. The HTLSTM model gives our algorithm the potential to fully consider the hierarchical information and long-term dependencies of language. We design the experiments on both English and Chinese corpus to evaluate our model on sentiment analysis task. And the results show that our model outperforms several existing state of the art approaches significantly.

Keywords: deep learning, hierarchical tree long short-term memory, sentence representation, sentiment analysis

Procedia PDF Downloads 349

Authors: Mariam Shah

Abstract:

Stereotypes are not generally considered to be an acceptable influence upon any decision making process, particularly those involving judicial decision making outcomes. Yet, we are confronted with an uncomfortable truth that stereotypes may be operating to influence judicial outcomes. Variances in sentencing outcomes are not easily explained away by criminological, psychological, or sociological theorem, but may be answered via qualitative research produced within the field of phenomenology. This paper will examine the current literature pertaining to the effect of stereotypes on the criminal justice system within the UK, and will also discuss what the implications are for stereotypical influences upon decision making in the criminal justice system. This paper will give particular focus to shoplifting offences dealt with in UK criminal courts, but this research has long reaching implications for the criminal process more generally.

Keywords: decision making, judicial decision making, phenomenology, shoplifting, stereotypes

Procedia PDF Downloads 337

Authors: Seema Vaidya

Abstract:

Mining association rule is a key issue in data mining. In any case, the standard models ignore the distinction among the exchanges, and the weighted association rule mining does not transform on databases with just binary attributes. This paper proposes a novel continuous example and executes a tree (FP-tree) structure, which is an increased prefix-tree structure for securing compacted, discriminating data about examples, and makes a fit FP-tree-based mining system, FP enhanced capacity algorithm is used, for mining the complete game plan of examples by illustration incessant development. Here, this paper handles the motivation behind making remarkable and weighted item sets, i.e. rare weighted item set mining issue. The two novel brightness measures are proposed for figuring the infrequent weighted item set mining issue. Also, the algorithm are handled which perform IWI which is more insignificant IWI mining. Moreover we utilized the rare item set for choice based structure. The general issue of the start of reliable definite rules is troublesome for the grounds that hypothetically no inciting technique with no other person can promise the rightness of influenced theories. In this way, this framework expects the disorder with the uncommon signs. Usage study demonstrates that proposed algorithm upgrades the structure which is successful and versatile for mining both long and short diagnostics rules. Structure upgrades aftereffects of foreseeing rare diseases of patient.

Keywords: association rule, data mining, IWI mining, infrequent item set, frequent pattern growth

Procedia PDF Downloads 400

Authors: Kamran Yazdanbakhsh, Somayeh Mahmoudi

Abstract:

Three reasons make good use of the decision theory in medicine: 1. Increased medical knowledge and their complexity makes it difficult treatment information effectively without resorting to sophisticated analytical methods, especially when it comes to detecting errors and identify opportunities for treatment from databases of large size. 2. There is a wide geographic variability of medical practice. In a context where medical costs are, at least in part, by the patient, these changes raise doubts about the relevance of the choices made by physicians. These differences are generally attributed to differences in estimates of probabilities of success of treatment involved, and differing assessments of the results on success or failure. Without explicit criteria for decision, it is difficult to identify precisely the sources of these variations in treatment. 3. Beyond the principle of informed consent, patients need to be involved in decision-making. For this, the decision process should be explained and broken down. A decision problem is to select the best option among a set of choices. The problem is what is meant by "best option ", or know what criteria guide the choice. The purpose of decision theory is to answer this question. The systematic use of decision models allows us to better understand the differences in medical practices, and facilitates the search for consensus. About this, there are three types of situations: situations certain, risky situations, and uncertain situations: 1. In certain situations, the consequence of each decision are certain. 2. In risky situations, every decision can have several consequences, the probability of each of these consequences is known. 3. In uncertain situations, each decision can have several consequences, the probability is not known. Our aim in this article is to show how decision theory can usefully be mobilized to meet the needs of physicians. The decision theory can make decisions more transparent: first, by clarifying the data systematically considered the problem and secondly by asking a few basic principles should guide the choice. Once the problem and clarified the decision theory provides operational tools to represent the available information and determine patient preferences, and thus assist the patient and doctor in their choices.

Keywords: decision making, medicine, treatment strategies, patient

Procedia PDF Downloads 579

Authors: Allaw Kamel, Al-Chami Leila

Abstract:

It is important to reduce the human influence on natural resources by identifying an appropriate land use. Moreover, it is essential to carry out the scientific land evaluation. Such kind of analysis allows identifying the main factors of agricultural production and enables decision makers to develop crop management in order to increase the land capability. The key is to match the type and intensity of land use with its natural capability. Therefore; in order to benefit from these areas and invest them to obtain good agricultural production, they must be organized and managed in full. Lebanon suffers from the unorganized agricultural use. We take south Lebanon as a study area, it is the most fertile ground and has a variety of crops. The study aims to identify and locate the most suitable area to cultivate thirteen type of permanent trees which are: apples, avocados, stone fruits in coastal regions and stone fruits in mountain regions, bananas, citrus, loquats, figs, pistachios, mangoes, olives, pomegranates, and grapes. Several geographical factors are taken as criterion for selection of the best location to cultivate. Soil, rainfall, PH, temperature, and elevation are main inputs to create the final map. Input data of each factor is managed, visualized and analyzed using Geographic Information System (GIS). Management GIS tools are implemented to produce input maps capable of identifying suitable areas related to each index. The combination of the different indices map generates the final output map of the suitable place to get the best permanent tree productivity. The output map is reclassified into three suitability classes: low, moderate, and high suitability. Results show different locations suitable for different kinds of trees. Results also reflect the importance of GIS in helping decision makers finding a most suitable location for every tree to get more productivity and a variety in crops.

Keywords: agricultural production, crop management, geographical factors, Geographic Information System, GIS, land capability, permanent trees, suitable location

Procedia PDF Downloads 142

Authors: Arwa Khannoussi, Alexandru-Liviu Olteanu, Pritesh Narayan, Catherine Dezan, Jean-Philippe Diguet, Patrick Meyer, Jacques Petit-Frere

Abstract:

Due to the nature of autonomous Unmanned Aerial Vehicles (UAV) missions, it is important that the decisions of a UAV stay consistent with the priorities of an operator, while at the same time allowing them to be easily audited and explained. We propose a multi-layer decision engine that integrates the operator (human) preferences by using the Multi-Criteria Decision Aiding (MCDA) methods. A software implementation of a UAV simulator and of the decision engine is presented to highlight the advantage of using such techniques on high-level decisions. We demonstrate that, with such a preference-based decision engine, the decisions of the UAV are compatible with the priorities of the operator, which in turn increases her/his confidence in its autonomous behavior.

Keywords: autonomous UAV, multi-criteria decision aiding, multi-layers decision engine, operator's preferences, traceable decisions, UAV simulation

Procedia PDF Downloads 256

Authors: Beema Akbar, Varun P. Gopi, V. Suresh Babu

Abstract:

Colon cancer causes the deaths of about half a million people every year. The common method of its detection is histopathological tissue analysis, it leads to tiredness and workload to the pathologist. A novel method is proposed that combines both structural and statistical pattern recognition used for the detection of colon cancer. This paper presents a comparison among the different classifiers such as Multilayer Perception (MLP), Sequential Minimal Optimization (SMO), Bayesian Logistic Regression (BLR) and k-star by using classification accuracy and error rate based on the percentage split method. The result shows that the best algorithm in WEKA is MLP classifier with an accuracy of 83.333% and kappa statistics is 0.625. The MLP classifier which has a lower error rate, will be preferred as more powerful classification capability.

Keywords: colon cancer, histopathological image, structural and statistical pattern recognition, multilayer perception

Procedia PDF Downloads 575

Authors: G. Sava, I. Popa, T. Sava, A. Ion, M. Ilie, C. Manailescu, A. Robu

Abstract:

At the RoAMS laboratory in Bucharest we have looked for a head-to-head meeting between AMS radiocarbon dating and dendrochronology dating, aiming to point out and explain any differences or similarities that might appear between their output results. As a subject of this investigation, we have fixed our attention on a sequence of tree rings spanning on a period of 700 years, starting with 1000 AD. The samples were collected from the northern Romanian territory within Moldavia region, and were provided by the ‘Marin Dracea - National Institute for Research and Development in Forestry’. All the 23 single ring wood samples were radiocarbon dated using alpha-cellulose extraction, followed by graphitization in an AGE3 installation. A wiggle matching procedure was applied to reduce the radiocarbon uncertainties for the calibrated ages. The results showed a good agreement on 3 out of 4 wood cores, the age-shifting of one of the wood cores being interpreted as an uncertain dendrochronology matching, which was further corrected.

Keywords: wiggle matching, tree-ring radiocarbon dating, dendrochronology, AMS radiocarbon dating, radiocarbon dating in Romania

Procedia PDF Downloads 183

Authors: Sofia F. Franco, Jacob Macdonald

Abstract:

This paper estimates the value of urban public street trees and their complementary and substitution value with other broader urban amenities and dis-amenities via the residential housing market. We estimate a lower bound value on a city’s tree amenities under instrumental variable and geographic regression discontinuity approaches with an application to Lisbon, Portugal. For completeness, we also explore how urban trees and in particular public street trees impact house prices across the city. Finally, we jointly analyze the planting and maintenance costs and benefits of urban street trees. The estimated value of all public trees in Lisbon is €8.84M. When considering specifically trees planted alongside roads and in public squares, the value is €6.06M or €126.64 per tree. This value is conditional on the distribution of trees in terms of their broader density, with higher effects coming from the overall greening of larger areas of the city compared to the greening of the direct neighborhood. Detrimental impacts are found when the number of trees is higher near street canyons, where they may exacerbate the stagnation of air pollution from traffic. Urban street trees also have important spillover benefits due to pollution mitigation around €6.21 million, or an additional €129.93 per tree. There are added benefits of €26.32 and €28.58 per tree in terms of flooding and heat mitigation, respectively. With significant resources and policies aimed at urban greening, the value obtained is shown to be important for discussions on the benefits of urban trees as compared to mitigation and abatement costs undertaken by a municipality.

Keywords: urban public goods, urban street trees, spatial boundary discontinuities, geospatial and remote sensing methods

Procedia PDF Downloads 178

Authors: Gopinath Rathod, Vinod Puranik

Abstract:

Sustainability is regarded as a key concept for survival in the competitive scenario. Industrial risk and diversification of risk type’s increases with industrial developments. In the context of sustainable manufacturing, the evaluation of risk is difficult because of the incomplete information and multiple indicators. Fuzzy Multiple Attribute Decision Method (FMADM) has been used with a three level hierarchical decision making model to evaluate aggregate risk for sustainable manufacturing projects. A case study has been presented to reflect the risk characteristics in sustainable manufacturing projects.

Keywords: sustainable manufacturing, decision making, aggregate risk, fuzzy logic, fuzzy multiple attribute decision method

Procedia PDF Downloads 521

Authors: Tatsuya Otoshi, Masayuki Murata

Abstract:

With the development and diversification of applications on the Internet, applications that require high responsiveness, such as video streaming, are becoming mainstream. Application responsiveness is not only a matter of communication delay but also a matter of time required to grasp changes in network conditions. The tradeoff between accuracy and measurement time is a challenge in network control. We people make countless decisions all the time, and our decisions seem to resolve tradeoffs between time and accuracy. When making decisions, people are known to make appropriate choices based on relatively small samples. Although there have been various studies on models of human decision-making, a model that integrates various cognitive biases, called ”quantum decision-making,” has recently attracted much attention. However, the modeling of small samples has not been examined much so far. In this paper, we extend the model of quantum decision-making to model decision-making with a small sample. In the proposed model, the state is updated by value-based probability amplitude amplification. By analytically obtaining a lower bound on the number of samples required for decision-making, we show that decision-making with a small number of samples is feasible.

Keywords: quantum decision making, small sample, MPEG-DASH, Grover's algorithm

Procedia PDF Downloads 80

Authors: Dasom Kim, William Xiu Shun Wong, Yoonjin Hyun, Donghoon Lee, Minji Paek, Sungho Byun, Namgyu Kim

Abstract:

There have been many researches on document classification for classifying voluminous documents automatically. Through document classification, we can assign a specific category to each unlabeled document on the basis of various machine learning algorithms. However, providing labeled documents manually requires considerable time and effort. To overcome the limitations, the semi-supervised learning which uses unlabeled document as well as labeled documents has been invented. However, traditional document classifiers, regardless of supervised or semi-supervised ones, cannot sufficiently explain the reason or the process of the classification. Thus, in this paper, we proposed a methodology to visualize major topics and class components of each document. We believe that our methodology for visualizing topics and classes of each document can enhance the reliability and explanatory power of document classifiers.

Keywords: data mining, document classifier, text mining, topic modeling

Procedia PDF Downloads 403

Authors: Jiahe Liu, Hongyang Yu, Miao Luo, Feng Qian

Abstract:

The widespread application of 3D human body reconstruction spans various fields. Smplify-X, an algorithm reliant on single-image input, employs three distinct body parameter templates, necessitating gender classification of individuals within the input image. Researchers employed a ResNet18 network to train a gender classifier within the Smplify-X framework, setting the threshold at 0.9, designating images falling below this threshold as having neutral gender. This model achieved 62.38% accurate predictions and 7.54% incorrect predictions. Our improvement involved refining the MobileNet network, resulting in a raised threshold of 0.97. Consequently, we attained 78.89% accurate predictions and a mere 0.2% incorrect predictions, markedly enhancing prediction precision and enabling more precise 3D human body reconstruction.

Keywords: SMPLX, mobileNet, gender classification, 3D human reconstruction

Procedia PDF Downloads 100

Authors: Kang Woobang

Abstract:

The base of the column is not only a support but also the embodiment of profound symbolism full of cosmic energy. Finding the full jars from which various energy emanate at the Notre Dame of Paris and Saint-Germain-des-Pres in France, the author was so shocked. As the column is cosmic tree, from the Full Jar full with cosmic energy emerges the cosmic tree composed of shaft and capital.

Keywords: full picher or jar, transcendental or supernatural birth from yonggi, yonggimun, yonggissak

Procedia PDF Downloads 413

Authors: Yassine Messaoud, Gordon Nigh, Faouzi Messaoud, Han Chen

Abstract:

The effect of climate change on tree growth in boreal and temperate forests has received increased interest in the context of global warming. However, most studies were conducted in small areas and with a limited number of tree species. Here, we examined the height growth responses of seventeen tree species to climate change in Western North America. 37009 stands from forest inventory databases in Canada and USA with varying establishment date were selected. Dominant and co-dominant trees from each stand were sampled to determine top tree height at 50 years breast height age. Height was related to historical mean annual and summer temperatures, annual and summer Palmer Drought Severity Index, tree establishment date, slope, aspect, soil fertility as determined by the rate of carbon organic matter decomposition (carbon/nitrogen), geographic locations (latitude, longitude, and elevation), species range (coastal, interior, and both ranges), shade tolerance and leaf form (needle leaves, deciduous needle leaves, and broadleaves). Climate change had mostly a positive effect on tree height growth. The results explained 62.4% of the height growth variance. Since 1880, height growth increase was greater for coastal, high shade tolerant, and broadleaf species. Height growth increased more on steep slopes and high soil fertility soils. Greater height growth was mostly observed at the leading range and upward. Conversely, some species showed the opposite pattern probably due to the increase of drought (coastal Mediterranean area), precipitation and cloudiness (Alaska and British Columbia) and peculiarity (higher latitudes-lower elevations and vice versa) of western North America topography. This study highlights the role of the species ecological amplitude and traits, and geographic locations as the main factors determining the growth response and its magnitude to the recent global climate change.

Keywords: Height growth, global climate change, species range, species characteristics, species ecological amplitude, geographic locations, western North America

Procedia PDF Downloads 187

Authors: I. Slim, H. Akkari, A. Ben Abdallah, I. Bhouri, M. Hedi Bedoui

Abstract:

Osteoporosis is a common disease characterized by low bone mass and deterioration of micro-architectural bone tissue, which provokes an increased risk of fracture. This work treats the texture characterization of trabecular bone radiographs. The aim was to analyze according to clinical research a group of 174 subjects: 87 osteoporotic patients (OP) with various bone fracture types and 87 control cases (CC). To characterize osteoporosis, Fractal and MultiFractal (MF) methods were applied to images for features (attributes) extraction. In order to improve the results, a new method of MF spectrum based on the q-stucture function calculation was proposed and a combination of Fractal and MF attributes was used. The Support Vector Machines (SVM) was applied as a classifier to distinguish between OP patients and CC subjects. The features fusion (fractal and MF) allowed a good discrimination between the two groups with an accuracy rate of 96.22%.

Keywords: fractal, micro-architecture analysis, multifractal, osteoporosis, SVM

Procedia PDF Downloads 393

Authors: Shivani Singh, Shantanu Nakhare, Kalyani Nair, Rohan Shetty

Abstract:

As social networking is growing at a rapid pace today it is vital that we work on improving its management. Research has shown that the content present in online social networks may have significant influence on impressionable minds. If such platforms are misused, it will lead to negative consequences. Detecting insults or inappropriate messages continues to be one of the most challenging aspects of Online Social Networks (OSNs) today. We address this problem through a Machine Learning Based Soft Text Classifier approach using Support Vector Machine algorithm. The proposed system acts as a screening mechanism the alerts the user about such messages. The messages are classified according to their subject matter and each comment is labeled for the presence of profanity and insults.

Keywords: machine learning, online social networks, soft text classifier, support vector machine

Procedia PDF Downloads 509

Authors: A. M. John-Otumu, M. M. Rahman, O. C. Nwokonkwo, M. C. Onuoha

Abstract:

Online social media networks have long served as a primary arena for group conversations, gossip, text-based information sharing and distribution. The use of natural language processing techniques for text classification and unbiased decision-making has not been far-fetched. Proper classification of this textual information in a given context has also been very difficult. As a result, we decided to conduct a systematic review of previous literature on sentiment classification and AI-based techniques that have been used in order to gain a better understanding of the process of designing and developing a robust and more accurate sentiment classifier that can correctly classify social media textual information of a given context between hate speech and inverted compliments with a high level of accuracy by assessing different artificial intelligence techniques. We evaluated over 250 articles from digital sources like ScienceDirect, ACM, Google Scholar, and IEEE Xplore and whittled down the number of research to 31. Findings revealed that Deep learning approaches such as CNN, RNN, BERT, and LSTM outperformed various machine learning techniques in terms of performance accuracy. A large dataset is also necessary for developing a robust sentiment classifier and can be obtained from places like Twitter, movie reviews, Kaggle, SST, and SemEval Task4. Hybrid Deep Learning techniques like CNN+LSTM, CNN+GRU, CNN+BERT outperformed single Deep Learning techniques and machine learning techniques. Python programming language outperformed Java programming language in terms of sentiment analyzer development due to its simplicity and AI-based library functionalities. Based on some of the important findings from this study, we made a recommendation for future research.

Keywords: artificial intelligence, natural language processing, sentiment analysis, social network, text

Procedia PDF Downloads 116

Authors: Dimitrios J. Dimitriou, Maria F. Sartzetaki, Anastasia Kalamakidou

Abstract:

Transport infrastructure assets are key components of the national asset portfolio. The decision to invest in a new infrastructure in transports could take from a few years to some decades. This is mainly because of the need to reserve and spent many capitals, the long payback period, the number of the stakeholders involved in the decision process and –many times- the investment and business risks are high. Decision makers and stakeholders need to define the framework and the outputs of the decision process taking into account the project characteristics, the business uncertainties, and the different expectations. Therefore, the decision assessment framework is an essential challenge linked with the key decision factors meet the stakeholder expectations highlighting project trade-offs, financial risks, business uncertainties and market limitations. This paper examines the decision process for new transport infrastructure projects in cross-border regions, where a wide range of stakeholders with different expectation is involved. According to a consequences analysis systemic approach, the relationship of transport infrastructure development, economic system development and stakeholder expectation is analysed. Adopting the on system of system methodological approach, the decision making the framework, variables, inputs and outputs are defined, highlighting the key shareholder’s role and expectations. The application provides the methodology outputs presenting the proposed decision framework for a strategic railway project in north Greece deals with the upgrade of the existing railway corridor connecting Greece, Turkey, and Bulgaria.

Keywords: system of system decision making, managing decisions for transport projects, decision support framework, defining decision process

Procedia PDF Downloads 309

Authors: A. Porshnev, A. Zaporozhtchuk

Abstract:

Motivation regarded as a key factor of labor turnover, is especially important for volunteers working on an altruistic basis in NGO. Despite the motivational letter, candidate selection depends on the impression of the selection committee, which can be subject to human bias. We expect that structured and unstructured information provided in motivation letters could be used to improve candidate selection procedures. In our paper, we perform qualitative and quantitative analysis of 2280 motivation letters, create logistic regression, and build a decision tree to improve selection procedures. Our analysis showed that motivation factors are significant and enable human resources department to forecast labor turnover and provide extra information to demographic, professional and timing questions. In spite of the average level of accuracy the model demonstrates the selection procedures of company of under consideration can be improved. We also discuss interrelation between answers to open and closed motivation questions, recommend changes in motivational letter templates to ensure more relevant information about applicants and further steps to create more accurate model.

Keywords: decision trees, logistic regression, model, motivational letter, non-governmental organization, retention, turnover

Procedia PDF Downloads 178

Authors: Amritpal Slaich, Mark Elshaw

Abstract:

This paper explores the development of a decision support framework for the introduction of business intelligence (BI) through operational research techniques for application by SMEs. Aligned with the goals of the new Midlands Enterprise Initiative of improving the skill levels of the Midlands workforce and addressing high levels of regional unemployment, we have developed a framework to increase the level of business intelligence used by SMEs to improve business decision-making. Many SMEs in the Midlands fail due to the lack of high quality decision making. Our framework outlines how universities can: engage with SMEs in the use of BI through operational research techniques; develop appropriate and easy to use Excel spreadsheet models; and make use of a process to allow SMEs to feedback their findings of the models. Future work will determine how well the framework performs in getting SMEs to apply BI to improve their decision-making performance.

Keywords: SMEs, decision support framework, business intelligence, operational research techniques

Procedia PDF Downloads 474

Authors: Fatima Akhmedova, Simon Liao

Abstract:

One of the most critical decision points in the design of a face recognition system is the choice of an appropriate face representation. Effective feature descriptors are expected to convey sufficient, invariant and non-redundant facial information. In this work, we propose a set of Hahn moments as a new approach for feature description. Hahn moments have been widely used in image analysis due to their invariance, non-redundancy and the ability to extract features either globally and locally. To assess the applicability of Hahn moments to Face Recognition we conduct two experiments on the Olivetti Research Laboratory (ORL) database and University of Notre-Dame (UND) X1 biometric collection. Fusion of the global features along with the features from local facial regions are used as an input for the conventional k-NN classifier. The method reaches an accuracy of 93% of correctly recognized subjects for the ORL database and 94% for the UND database.

Keywords: face recognition, Hahn moments, recognition-by-parts, time-lapse

Procedia PDF Downloads 377

Authors: Siqi Qiu, Yijian Zheng, Xin Guo Ming

Abstract:

In many reliability and risk analysis, failures of components are supposed to be independent. However, in reality, the ignorance of failure dependencies among components may render the results of reliability and risk analysis incorrect. There are two principal ways to incorporate failure dependencies in system reliability and risk analysis: implicit and explicit methods. In the implicit method, failure dependencies can be modeled by joint probabilities, correlation values or conditional probabilities. In the explicit method, certain types of dependencies can be modeled in a fault tree as mutually independent basic events for specific component failures. In this paper, explicit and implicit methods based on BDD will be proposed to evaluate the reliability of systems considering failure dependencies. The obtained results prove the equivalence of the proposed implicit and explicit methods. It is found that the consideration of failure dependencies decreases the reliability of systems. This observation is intuitive, because more components fail due to failure dependencies. The consideration of failure dependencies helps designers to reduce the dependencies between components during the design phase to make the system more reliable.

Keywords: reliability assessment, risk assessment, failure dependencies, binary decision diagram

Procedia PDF Downloads 472