21 |
應用文字探勘文件分類分群技術於股價走勢預測之研究─以台灣股票市場為例 / A Study of Stock Price Prediction with Text Mining, Classification and Clustering Techniques in Taiwan Stock Market薛弘業, Hsueh, Hung Yeh Unknown Date (has links)
本研究欲探究個股新聞影響台灣股票市場之關係,透過蒐集宏達電、台積電與鴻海等三間上市公司從2012年6月至2013年5月的歷史交易資料和個股新聞,使用文字探勘技術找出各新聞內容的特徵,再透過歷史資料、技術分析指標與kNN和2-way kNN演算法將新聞先做分類後分群,建立預測模型,分析新聞對股價漲跌的影響與程度,以及漲跌幅度較高之群集與股價漲跌和轉折的關係。
研究結果發現,加入技術分析指標後能夠提升分類的準確率,而漲跌類別內的分群能夠界定各群集與股價漲跌之間的關係,且漲跌幅度較高之群集的分析則能大幅提升投資準確率至80%左右,而股價轉折點之預測則能提供一個明確的投資進場時間點,並確保當投資人依照此預測模型的結果進行7交易日投資時,可以在風險極低的前提下,穩當且迅速的獲取2.82%至22.03%不等的投資報酬。 / This study investigated the relation that the stock news effect on Taiwan Stock Market. Through collected the historical transaction data and stock news from July, 2012 to May, 2013, and use text mining、kNN Classification and 2-Way kNN Clustering technique analyzing the stock news, build a forecast model to analyze the degree of news effect on the stock price, and find the relation between the cluster which has great degree and the reversal points of stock price.
The result shows that using the change range and Technical Indicator rise classification’s accuracy, and clustering in the ”up” group and “down” group can identify the range stock price move, and rise the invested accuracy up to about 80 percent. The forecast of reversal points of stock price offers a specific time to invest, and insure the investors who execute a 7 trading day investment depend on this model can get 2.82 to 22.03 percent return reliably and quickly with low risk.
|
22 |
An Edge-Based Algorithm for Spatial Query Processing in Real-Life Road NetworksWu, Xu-Lun 14 July 2011 (has links)
Due to wireless communication technologies, positioning technologies, and mobile computing develop quickly, mobile services are becoming practical and important on
big spatiotemporal databases management. Mobile service users move only inside a spatial network, e.g. a road network. They often issue the K Nearest Neighbor (KNN) query to obtain data objects reachable through the road network. The challenge problem of mobile services is how to efficiently answer the data objects which user interest to the corresponding mobile users. Therefore, how to effectively modeling road networks, effectively indexing, and querying on the road networks have become a popular topic. Lu et. al. have proposed a road network model that captures the real-life road networks better than previous models. Then, based on their model, they have proposed a RNG (Road Network Grid) index for speeding up the KNN
query on real-life road networks. The RNG index structure is a quad-tree structure and a point-based structure. However, in their model, they divide the double track road which U-turn is allowed at some parts. This modeling does not capture the real-life road networks accurately. Since they divide the road, this makes the number of points of the graph increase. The number of times of partitioning the graph increases. It increases the execution time of constructing the index structure. The format of the leaf node of the RNG index makes the search time increase. Moreover, the query processing on the RNG index structure has to visit the root repeatedly. This condition makes the search time increase. Therefore, in this thesis, we propose a network model that captures the real-life road networks. We do not have to divide the real-life roads when we map the real-life roads into graph. We map the real-life road networks into graph directly. Then, based on our network model, we propose an EBNA (Edge-Based Nine-Area tree) index structure to make the search time of obtaining the interest edge information quickly. The EBNA index structure is an edge-based index structure. We store all of the edge information on the leaf node. We can obtain the edge information directly. Each edge information entry has a
pointer point to link edges. Links of each edge entry consist a graph. This graph makes the KNN query processing visit the root only one time. From our simulation result, we show that the performance of constructing the EBNA index is better than constructing the RNG index and the performance of the KNN query processing by using EBNA index is better than the KNN query processing by using RNG index.
|
23 |
KNN Query Processing in Wireless Sensor and Robot NetworksXie, Wei 28 February 2014 (has links)
In Wireless Sensor and Robot Networks (WSRNs), static sensors report event information to one of the robots. In the k nearest neighbour query processing problem in WSRNs, the robot receives event report needs to find exact k nearest robots (KNN) to react to the event, among those connected to it. We are interested in localized solutions, which avoid message flooding to the whole network. Several existing methods restrict the search within a predetermined boundary. Some network density-based estimation algorithms were proposed but they either result in large message transmission or require the density information of the whole network in advance which is complex to implement and lacks robustness. Algorithms with tree structures lead to the excessive energy consumption and large latency caused by structural construction. Itinerary based approaches generate large latency or unsatisfactory accuracy. In this thesis, we propose a new method to estimate a search boundary, which is a circle centred at the query point. Two algorithms are presented to disseminate the message to robots of interest and aggregate their data (e.g. the distance to query point). Multiple Auction Aggregation (MAA) is an algorithm based on auction protocol, with multiple copies of query message being disseminated into the network to get the best bidding from each robot. Partial Depth First Search (PDFS) attempts to traverse all the robots of interest with a query message to gather the data by depth first search. This thesis also optimizes a traditional itinerary-based KNN query processing method called IKNN and compares this algorithm with our proposed MAA and PDFS algorithms. The experimental results followed indicate that the overall performance of MAA and PDFS outweighs IKNN in WSRNs.
|
24 |
Classificação de mangas Tommy Atkins y-irradiadas: Um modelo metabolômicoSANTOS, Maria de Jesus Lessa 31 January 2014 (has links)
Submitted by Danielle Karla Martins Silva (danielle.martins@ufpe.br) on 2015-03-13T13:23:18Z
No. of bitstreams: 2
DISSERTAÇÃO Maria de Jesus Santos.pdf: 2877099 bytes, checksum: 04af70722e8d7edebcc4a65d2877869b (MD5)
license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) / Made available in DSpace on 2015-03-13T13:23:18Z (GMT). No. of bitstreams: 2
DISSERTAÇÃO Maria de Jesus Santos.pdf: 2877099 bytes, checksum: 04af70722e8d7edebcc4a65d2877869b (MD5)
license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5)
Previous issue date: 2014 / Neste trabalho foram investigadas as composições dos voláteis a partir de mangas da cultivar Tommy Atkins expostas à radiação gama na dose de 0,5 kGy quando comparadas à composição de voláteis obtidos a partir de mangas que não passaram por este tratamento fitossanitário. O objetivo foi construir um modelo metabolômico para classificar as mangas através de modelo não invasivo. Foram analisadas 80 amostras classificadas com grau de maturação entre 4 e 5, segundo classificação da Embrapa. Os voláteis foram coletados após 18 dias de armazenamento sob temperatura de 12°C, usando um sistema Headspace Dinâmico (HD) e submetidos à corrida cromatográfica em fase gasosa seguida de detecção por espectrometria de massas (GC/MS). Os compostos foram identificados a partir da determinação do Índice de Retenção Van den Dool and Kratz e do espectro de massas, que foram comparados aos descritos na biblioteca de espectros do ADAMS. Foram identificados 16 compostos já mencionados na literatura e classificados como terpenos (mono e sesquiterpenos) e ésteres. Entre os terpenos, o α-Pineno e o 3-Careno foram os majoritários tanto para as mangas irradiadas, como para as não irradiadas. Após a identificação dos mesmos, os cromatogramas foram utilizados para a construção de uma matriz para tratamento estatístico, o qual foi realizado utilizando a plataforma online MetaboAnalyst 2.0 e o software “R Program”. As ferramentas de estatística multivariada utilizadas foram a PCA (Análise de Componentes Principais), PLS-DA (Análise Discriminante e Regressão por Mínimos Quadrados Parciais) e KNN (K-nearest neighbor). A PCA não apresentou um resultado satisfatório para a discriminação entre as mangas irradiadas e não irradiadas. Na PLS-DA, treze compostos foram responsáveis pela discriminação entre as mangas da cultivar Tommy Atkins irradiadas e não irradiadas, com destaque para o Octanoato de Etila, o α-Felandreno e o Germacreno-D. O KNN também indicou que os teores de Octanoato de Etila, α-Felandreno e Germacreno-D são responsáveis pela discriminação entre as mangas irradiadas e não irradiadas. No entanto, a acurácia observada na classificação utilizando KNN foi maior que a observada utilizando PLS-DA. No modelo construído com KNN, o teste de validação cruzada indicou acurácia igual a 81% contra 55% da observada para o modelo construído utilizando PLS-DA. Esse resultado garante um modelo metabolômico que é capaz de classificar as amostras de mangas da cultivar Tommy Atkins que foram expostas, ou não, à radiação gama para fins fitossanitários.
|
25 |
Predictive models for chronic renal disease using decision trees, naïve bayes and case-based methodsKhan, Saqib Hussain January 2010 (has links)
Data mining can be used in healthcare industry to “mine” clinical data to discover hidden information for intelligent and affective decision making. Discovery of hidden patterns and relationships often goes intact, yet advanced data mining techniques can be helpful as remedy to this scenario. This thesis mainly deals with Intelligent Prediction of Chronic Renal Disease (IPCRD). Data covers blood, urine test, and external symptoms applied to predict chronic renal disease. Data from the database is initially transformed to Weka (3.6) and Chi-Square method is used for features section. After normalizing data, three classifiers were applied and efficiency of output is evaluated. Mainly, three classifiers are analyzed: Decision Tree, Naïve Bayes, K-Nearest Neighbour algorithm. Results show that each technique has its unique strength in realizing the objectives of the defined mining goals. Efficiency of Decision Tree and KNN was almost same but Naïve Bayes proved a comparative edge over others. Further sensitivity and specificity tests are used as statistical measures to examine the performance of a binary classification. Sensitivity (also called recall rate in some fields) measures the proportion of actual positives which are correctly identified while Specificity measures the proportion of negatives which are correctly identified. CRISP-DM methodology is applied to build the mining models. It consists of six major phases: business understanding, data understanding, data preparation, modeling, evaluation, and deployment.
|
26 |
KNN Query Processing in Wireless Sensor and Robot NetworksXie, Wei January 2014 (has links)
In Wireless Sensor and Robot Networks (WSRNs), static sensors report event information to one of the robots. In the k nearest neighbour query processing problem in WSRNs, the robot receives event report needs to find exact k nearest robots (KNN) to react to the event, among those connected to it. We are interested in localized solutions, which avoid message flooding to the whole network. Several existing methods restrict the search within a predetermined boundary. Some network density-based estimation algorithms were proposed but they either result in large message transmission or require the density information of the whole network in advance which is complex to implement and lacks robustness. Algorithms with tree structures lead to the excessive energy consumption and large latency caused by structural construction. Itinerary based approaches generate large latency or unsatisfactory accuracy. In this thesis, we propose a new method to estimate a search boundary, which is a circle centred at the query point. Two algorithms are presented to disseminate the message to robots of interest and aggregate their data (e.g. the distance to query point). Multiple Auction Aggregation (MAA) is an algorithm based on auction protocol, with multiple copies of query message being disseminated into the network to get the best bidding from each robot. Partial Depth First Search (PDFS) attempts to traverse all the robots of interest with a query message to gather the data by depth first search. This thesis also optimizes a traditional itinerary-based KNN query processing method called IKNN and compares this algorithm with our proposed MAA and PDFS algorithms. The experimental results followed indicate that the overall performance of MAA and PDFS outweighs IKNN in WSRNs.
|
27 |
Robustní detekce klíčových slov v řečovém signálu / Robust detection of keywords in speech signalVrba, Václav January 2014 (has links)
The master thesis is divided into two parts theoretical and practical. The theoretical part is focused on methods of analysis and detection of speech signals. In the practical part the system for isolated word recognition was created in Matlab. The system is speaker independent separately for men and women. Also two speech databases were created for further use in the aircraft cockpit. Tests and evaluations were performed even with added noise.
|
28 |
Consensus-Oriented Cloud Additive Manufacturing Based on Blockchain Technology: An Exploratory Study on System Operation Efficiency and SecurityZhu, Xiaobao 16 June 2020 (has links)
No description available.
|
29 |
On Gender Identification Using the Smile DynamicsAl-dahoud, Ahmad, Ugail, Hassan January 2017 (has links)
No / Gender classification has multiple applications including, but not limited to, face perception, age, ethnicity and identity analysis, video surveillance and smart human computer interaction. The majority of computer based gender classification algorithms analyse the appearance of facial features predominantly based on the texture of the static image of the face. In this paper, we propose a novel algorithm for gender classification using the smile dynamics without resorting to the use of any facial texture information. Our experiments suggest that this method has great potential for finding indicators of gender dimorphism. Our approach was tested on two databases, namely the CK+ and the MUG, consisting of a total of 80 subjects. As a result, using the KNN algorithm along with 10-fold cross validation, we achieve an accurate classification rate of 80% for gender simply based on the dynamics of a person's smile.
|
30 |
Intrusion Detection using Bit Timing Characteristics for CAN BusPatel, Chitvan Kirit 19 July 2019 (has links)
In today's world, most automobiles use Controller Area Network (CAN) bus for communication between various Electronic Control Units (ECUs), also called nodes on the CAN bus. Each ECU on the CAN bus is a microcontroller that sends a unique identifier used for node identification. It is possible to spoof node A by sending the same identifier through node B and thereby control node A. Thus, a hacker can control the steering using the car's internal lights and render it ineffective or misuse them. In order to combat this, we try to fingerprint each node by identifying its identifier's unique bit timing characteristics. To that extent, bit timing characteristics used are the Time of Flight (TOF) intervals between successive rising edges of identifier bits, for an ECU. Similarly, other characteristics such as TOF between successive falling edges of the CAN bus node identifier can also be used for node classification.
In order to measure these TOFs, we use a device called Time-to-Digital Convertor, which essentially triggers a ring oscillator to measure time values between rising/falling edges of a signal, to the order of picosecond accuracy. These timing values are used as features into the K-nearest neighbors (KNN) classifier algorithm. Once the classifier is trained, it can be used to predict a new timing value into a particular node category, which if different from the expected category is a sign of compromise or intrusion. It is seen that we achieve 95% accuracy of correctly predicting the compromised node under simulation tests. Thereafter, the thesis deals with experimentally predicting an intrusion in the CAN bus system utilizing EPOS Studio CAN bus position controller for Maxon motors. The clock timings being extremely accurate leads to the conclusion that employment of better statistical techniques for node characterization is needed for intrusion detection, which is outside the scope of this work. / Master of Science / In today’s world, most automobiles use Controller Area Network (CAN) bus for communication between various Electronic Control Units (ECUs), also called nodes on the CAN bus. These nodes can range from car headlights, radio, doors, internal lights to brakes, steering, throttle and much more. Each node on the CAN bus is a microcontroller which controls its proper operation. This also means that if a node is compromised using external hardware or a piece of software, it could be quite risky. Thus, a hacker can control the steering using the car’s internal lights and render it ineffective or misuse them. In order to combat this, we try to fingerprint each node by identifying its unique time domain characteristics. These characteristics can be the Time of Flight (TOF) measurement values between successive rising or falling edges of a node’s unique identifier, using an instrument called a Time-to-Digital convertor. Furthermore, these TOF values are used as features for the K-nearest neighbor (KNN) classifier machine learning algorithm, which uniquely identifies signals coming from any of the fingerprinted nodes, thereby raising a flag if a message comes from an unidentified node. In addition, experimental data is obtained for node identifiers on the CAN bus, in digital form, and passed into a neural network (NN) for training the classifier. We achieve an 95% and 70% prediction accuracy for the KNN and NN classifiers respectively.
|
Page generated in 0.082 seconds