Matching Items (16)
Filtering by
- All Subjects: Machine Learning
- Genre: Doctoral Dissertation
- Creators: Davulcu, Hasan
- Member of: ASU Electronic Theses and Dissertations
- Resource Type: Text
Description
Multidimensional data have various representations. Thanks to their simplicity in modeling multidimensional data and the availability of various mathematical tools (such as tensor decompositions) that support multi-aspect analysis of such data, tensors are increasingly being used in many application domains including scientific data management, sensor data management, and social network data analysis. Relational model, on the other hand, enables semantic manipulation of data using relational operators, such as projection, selection, Cartesian-product, and set operators. For many multidimensional data applications, tensor operations as well as relational operations need to be supported throughout the data life cycle. In this thesis, we introduce a tensor-based relational data model (TRM), which enables both tensor- based data analysis and relational manipulations of multidimensional data, and define tensor-relational operations on this model. Then we introduce a tensor-relational data management system, so called, TensorDB. TensorDB is based on TRM, which brings together relational algebraic operations (for data manipulation and integration) and tensor algebraic operations (for data analysis). We develop optimization strategies for tensor-relational operations in both in-memory and in-database TensorDB. The goal of the TRM and TensorDB is to serve as a single environment that supports the entire life cycle of data; that is, data can be manipulated, integrated, processed, and analyzed.
ContributorsKim, Mijung (Author) / Candan, K. Selcuk (Thesis advisor) / Davulcu, Hasan (Committee member) / Sundaram, Hari (Committee member) / Ye, Jieping (Committee member) / Arizona State University (Publisher)
Created2014
Description
Advances in data collection technologies have made it cost-effective to obtain heterogeneous data from multiple data sources. Very often, the data are of very high dimension and feature selection is preferred in order to reduce noise, save computational cost and learn interpretable models. Due to the multi-modality nature of heterogeneous data, it is interesting to design efficient machine learning models that are capable of performing variable selection and feature group (data source) selection simultaneously (a.k.a bi-level selection). In this thesis, I carry out research along this direction with a particular focus on designing efficient optimization algorithms. I start with a unified bi-level learning model that contains several existing feature selection models as special cases. Then the proposed model is further extended to tackle the block-wise missing data, one of the major challenges in the diagnosis of Alzheimer's Disease (AD). Moreover, I propose a novel interpretable sparse group feature selection model that greatly facilitates the procedure of parameter tuning and model selection. Last but not least, I show that by solving the sparse group hard thresholding problem directly, the sparse group feature selection model can be further improved in terms of both algorithmic complexity and efficiency. Promising results are demonstrated in the extensive evaluation on multiple real-world data sets.
ContributorsXiang, Shuo (Author) / Ye, Jieping (Thesis advisor) / Mittelmann, Hans D (Committee member) / Davulcu, Hasan (Committee member) / He, Jingrui (Committee member) / Arizona State University (Publisher)
Created2014
Description
With the rise of social media, hundreds of millions of people spend countless hours all over the globe on social media to connect, interact, share, and create user-generated data. This rich environment provides tremendous opportunities for many different players to easily and effectively reach out to people, interact with them, influence them, or get their opinions. There are two pieces of information that attract most attention on social media sites, including user preferences and interactions. Businesses and organizations use this information to better understand and therefore provide customized services to social media users. This data can be used for different purposes such as, targeted advertisement, product recommendation, or even opinion mining. Social media sites use this information to better serve their users.
Despite the importance of personal information, in many cases people do not reveal this information to the public. Predicting the hidden or missing information is a common response to this challenge. In this thesis, we address the problem of predicting user attributes and future or missing links using an egocentric approach. The current research proposes novel concepts and approaches to better understand social media users in twofold including, a) their attributes, preferences, and interests, and b) their future or missing connections and interactions. More specifically, the contributions of this dissertation are (1) proposing a framework to study social media users through their attributes and link information, (2) proposing a scalable algorithm to predict user preferences; and (3) proposing a novel approach to predict attributes and links with limited information. The proposed algorithms use an egocentric approach to improve the state of the art algorithms in two directions. First by improving the prediction accuracy, and second, by increasing the scalability of the algorithms.
Despite the importance of personal information, in many cases people do not reveal this information to the public. Predicting the hidden or missing information is a common response to this challenge. In this thesis, we address the problem of predicting user attributes and future or missing links using an egocentric approach. The current research proposes novel concepts and approaches to better understand social media users in twofold including, a) their attributes, preferences, and interests, and b) their future or missing connections and interactions. More specifically, the contributions of this dissertation are (1) proposing a framework to study social media users through their attributes and link information, (2) proposing a scalable algorithm to predict user preferences; and (3) proposing a novel approach to predict attributes and links with limited information. The proposed algorithms use an egocentric approach to improve the state of the art algorithms in two directions. First by improving the prediction accuracy, and second, by increasing the scalability of the algorithms.
ContributorsAbbasi, Mohammad Ali, 1975- (Author) / Liu, Huan (Thesis advisor) / Davulcu, Hasan (Committee member) / Ye, Jieping (Committee member) / Agarwal, Nitin (Committee member) / Arizona State University (Publisher)
Created2014
Description
The subliminal impact of framing of social, political and environmental issues such as climate change has been studied for decades in political science and communications research. Media framing offers an “interpretative package" for average citizens on how to make sense of climate change and its consequences to their livelihoods, how to deal with its negative impacts, and which mitigation or adaptation policies to support. A line of related work has used bag of words and word-level features to detect frames automatically in text. Such works face limitations since standard keyword based features may not generalize well to accommodate surface variations in text when different keywords are used for similar concepts.
This thesis develops a unique type of textual features that generalize triplets extracted from text, by clustering them into high-level concepts. These concepts are utilized as features to detect frames in text. Compared to uni-gram and bi-gram based models, classification and clustering using generalized concepts yield better discriminating features and a higher classification accuracy with a 12% boost (i.e. from 74% to 83% F-measure) and 0.91 clustering purity for Frame/Non-Frame detection.
The automatic discovery of complex causal chains among interlinked events and their participating actors has not yet been thoroughly studied. Previous studies related to extracting causal relationships from text were based on laborious and incomplete hand-developed lists of explicit causal verbs, such as “causes" and “results in." Such approaches result in limited recall because standard causal verbs may not generalize well to accommodate surface variations in texts when different keywords and phrases are used to express similar causal effects. Therefore, I present a system that utilizes generalized concepts to extract causal relationships. The proposed algorithms overcome surface variations in written expressions of causal relationships and discover the domino effects between climate events and human security. This semi-supervised approach alleviates the need for labor intensive keyword list development and annotated datasets. Experimental evaluations by domain experts achieve an average precision of 82%. Qualitative assessments of causal chains show that results are consistent with the 2014 IPCC report illuminating causal mechanisms underlying the linkages between climatic stresses and social instability.
This thesis develops a unique type of textual features that generalize triplets extracted from text, by clustering them into high-level concepts. These concepts are utilized as features to detect frames in text. Compared to uni-gram and bi-gram based models, classification and clustering using generalized concepts yield better discriminating features and a higher classification accuracy with a 12% boost (i.e. from 74% to 83% F-measure) and 0.91 clustering purity for Frame/Non-Frame detection.
The automatic discovery of complex causal chains among interlinked events and their participating actors has not yet been thoroughly studied. Previous studies related to extracting causal relationships from text were based on laborious and incomplete hand-developed lists of explicit causal verbs, such as “causes" and “results in." Such approaches result in limited recall because standard causal verbs may not generalize well to accommodate surface variations in texts when different keywords and phrases are used to express similar causal effects. Therefore, I present a system that utilizes generalized concepts to extract causal relationships. The proposed algorithms overcome surface variations in written expressions of causal relationships and discover the domino effects between climate events and human security. This semi-supervised approach alleviates the need for labor intensive keyword list development and annotated datasets. Experimental evaluations by domain experts achieve an average precision of 82%. Qualitative assessments of causal chains show that results are consistent with the 2014 IPCC report illuminating causal mechanisms underlying the linkages between climatic stresses and social instability.
ContributorsAlashri, Saud (Author) / Davulcu, Hasan (Thesis advisor) / Desouza, Kevin C. (Committee member) / Maciejewski, Ross (Committee member) / Hsiao, Sharon (Committee member) / Arizona State University (Publisher)
Created2018
Description
The performance of most of the visual computing tasks depends on the quality of the features extracted from the raw data. Insightful feature representation increases the performance of many learning algorithms by exposing the underlying explanatory factors of the output for the unobserved input. A good representation should also handle anomalies in the data such as missing samples and noisy input caused by the undesired, external factors of variation. It should also reduce the data redundancy. Over the years, many feature extraction processes have been invented to produce good representations of raw images and videos.
The feature extraction processes can be categorized into three groups. The first group contains processes that are hand-crafted for a specific task. Hand-engineering features requires the knowledge of domain experts and manual labor. However, the feature extraction process is interpretable and explainable. Next group contains the latent-feature extraction processes. While the original feature lies in a high-dimensional space, the relevant factors for a task often lie on a lower dimensional manifold. The latent-feature extraction employs hidden variables to expose the underlying data properties that cannot be directly measured from the input. Latent features seek a specific structure such as sparsity or low-rank into the derived representation through sophisticated optimization techniques. The last category is that of deep features. These are obtained by passing raw input data with minimal pre-processing through a deep network. Its parameters are computed by iteratively minimizing a task-based loss.
In this dissertation, I present four pieces of work where I create and learn suitable data representations. The first task employs hand-crafted features to perform clinically-relevant retrieval of diabetic retinopathy images. The second task uses latent features to perform content-adaptive image enhancement. The third task ranks a pair of images based on their aestheticism. The goal of the last task is to capture localized image artifacts in small datasets with patch-level labels. For both these tasks, I propose novel deep architectures and show significant improvement over the previous state-of-art approaches. A suitable combination of feature representations augmented with an appropriate learning approach can increase performance for most visual computing tasks.
The feature extraction processes can be categorized into three groups. The first group contains processes that are hand-crafted for a specific task. Hand-engineering features requires the knowledge of domain experts and manual labor. However, the feature extraction process is interpretable and explainable. Next group contains the latent-feature extraction processes. While the original feature lies in a high-dimensional space, the relevant factors for a task often lie on a lower dimensional manifold. The latent-feature extraction employs hidden variables to expose the underlying data properties that cannot be directly measured from the input. Latent features seek a specific structure such as sparsity or low-rank into the derived representation through sophisticated optimization techniques. The last category is that of deep features. These are obtained by passing raw input data with minimal pre-processing through a deep network. Its parameters are computed by iteratively minimizing a task-based loss.
In this dissertation, I present four pieces of work where I create and learn suitable data representations. The first task employs hand-crafted features to perform clinically-relevant retrieval of diabetic retinopathy images. The second task uses latent features to perform content-adaptive image enhancement. The third task ranks a pair of images based on their aestheticism. The goal of the last task is to capture localized image artifacts in small datasets with patch-level labels. For both these tasks, I propose novel deep architectures and show significant improvement over the previous state-of-art approaches. A suitable combination of feature representations augmented with an appropriate learning approach can increase performance for most visual computing tasks.
ContributorsChandakkar, Parag Shridhar (Author) / Li, Baoxin (Thesis advisor) / Yang, Yezhou (Committee member) / Turaga, Pavan (Committee member) / Davulcu, Hasan (Committee member) / Arizona State University (Publisher)
Created2017
Description
This research start utilizing an efficient sparse inverse covariance matrix (precision matrix) estimation technique to identify a set of highly correlated discriminative perspectives between radical and counter-radical groups. A ranking system has been developed that utilizes ranked perspectives to map Islamic organizations on a set of socio-cultural, political and behavioral scales based on their web site corpus. Simultaneously, a gold standard ranking of these organizations was created through domain experts and compute expert-to-expert agreements and present experimental results comparing the performance of the QUIC based scaling system to another baseline method for organizations. The QUIC based algorithm not only outperforms the baseline methods, but it is also the only system that consistently performs at area expert-level accuracies for all scales. Also, a multi-scale ideological model has been developed and it investigates the correlates of Islamic extremism in Indonesia, Nigeria and UK. This analysis demonstrate that violence does not correlate strongly with broad Muslim theological or sectarian orientations; it shows that religious diversity intolerance is the only consistent and statistically significant ideological correlate of Islamic extremism in these countries, alongside desire for political change in UK and Indonesia, and social change in Nigeria. Next, dynamic issues and communities tracking system based on NMF(Non-negative Matrix Factorization) co-clustering algorithm has been built to better understand the dynamics of virtual communities. The system used between Iran and Saudi Arabia to build and apply a multi-party agent-based model that can demonstrate the role of wedges and spoilers in a complex environment where coalitions are dynamic. Lastly, a visual intelligence platform for tracking the diffusion of online social movements has been developed called LookingGlass to track the geographical footprint, shifting positions and flows of individuals, topics and perspectives between groups. The algorithm utilize large amounts of text collected from a wide variety of organizations’ media outlets to discover their hotly debated topics, and their discriminative perspectives voiced by opposing camps organized into multiple scales. Discriminating perspectives is utilized to classify and map individual Tweeter’s message content to social movements based on the perspectives expressed in their tweets.
ContributorsKim, Nyunsu (Author) / Davulcu, Hasan (Thesis advisor) / Sen, Arunabha (Committee member) / Hsiao, Sharon (Committee member) / Corman, Steven (Committee member) / Arizona State University (Publisher)
Created2018
Description
Cyber-systems and networks are the target of different types of cyber-threats and attacks, which are becoming more common, sophisticated, and damaging. Those attacks can vary in the way they are performed. However, there are similar strategies
and tactics often used because they are time-proven to be effective. The motivations behind cyber-attacks play an important role in designating how attackers plan and proceed to achieve their goals. Generally, there are three categories of motivation
are: political, economical, and socio-cultural motivations. These indicate that to defend against possible attacks in an enterprise environment, it is necessary to consider what makes such an enterprise environment a target. That said, we can understand
what threats to consider and how to deploy the right defense system. In other words, detecting an attack depends on the defenders having a clear understanding of why they become targets and what possible attacks they should expect. For instance,
attackers may preform Denial of Service (DoS), or even worse Distributed Denial of Service (DDoS), with intention to cause damage to targeted organizations and prevent legitimate users from accessing their services. However, in some cases, attackers are very skilled and try to hide in a system undetected for a long period of time with the incentive to steal and collect data rather than causing damages.
Nowadays, not only the variety of attack types and the way they are launched are important. However, advancement in technology is another factor to consider. Over the last decades, we have experienced various new technologies. Obviously, in the beginning, new technologies will have their own limitations before they stand out. There are a number of related technical areas whose understanding is still less than satisfactory, and in which long-term research is needed. On the other hand, these new technologies can boost the advancement of deploying security solutions and countermeasures when they are carefully adapted. That said, Software Defined Networking i(SDN), its related security threats and solutions, and its adaption in enterprise environments bring us new chances to enhance our security solutions. To reach the optimal level of deploying SDN technology in enterprise environments, it is important to consider re-evaluating current deployed security solutions in traditional networks before deploying them to SDN-based infrastructures. Although DDoS attacks are a bit sinister, there are other types of cyber-threats that are very harmful, sophisticated, and intelligent. Thus, current security defense solutions to detect DDoS cannot detect them. These kinds of attacks are complex, persistent, and stealthy, also referred to Advanced Persistent Threats (APTs) which often leverage the bot control and remotely access valuable information. APT uses multiple stages to break into a network. APT is a sort of unseen, continuous and long-term penetrative network and attackers can bypass the existing security detection systems. It can modify and steal the sensitive data as well as specifically cause physical damage the target system. In this dissertation, two cyber-attack motivations are considered: sabotage, where the motive is the destruction; and information theft, where attackers aim to acquire invaluable information (customer info, business information, etc). I deal with two types of attacks (DDoS attacks and APT attacks) where DDoS attacks are classified under sabotage motivation category, and the APT attacks are classified under information theft motivation category. To detect and mitigate each of these attacks, I utilize the ease of programmability in SDN and its great platform for implementation, dynamic topology changes, decentralized network management, and ease of deploying security countermeasures.
and tactics often used because they are time-proven to be effective. The motivations behind cyber-attacks play an important role in designating how attackers plan and proceed to achieve their goals. Generally, there are three categories of motivation
are: political, economical, and socio-cultural motivations. These indicate that to defend against possible attacks in an enterprise environment, it is necessary to consider what makes such an enterprise environment a target. That said, we can understand
what threats to consider and how to deploy the right defense system. In other words, detecting an attack depends on the defenders having a clear understanding of why they become targets and what possible attacks they should expect. For instance,
attackers may preform Denial of Service (DoS), or even worse Distributed Denial of Service (DDoS), with intention to cause damage to targeted organizations and prevent legitimate users from accessing their services. However, in some cases, attackers are very skilled and try to hide in a system undetected for a long period of time with the incentive to steal and collect data rather than causing damages.
Nowadays, not only the variety of attack types and the way they are launched are important. However, advancement in technology is another factor to consider. Over the last decades, we have experienced various new technologies. Obviously, in the beginning, new technologies will have their own limitations before they stand out. There are a number of related technical areas whose understanding is still less than satisfactory, and in which long-term research is needed. On the other hand, these new technologies can boost the advancement of deploying security solutions and countermeasures when they are carefully adapted. That said, Software Defined Networking i(SDN), its related security threats and solutions, and its adaption in enterprise environments bring us new chances to enhance our security solutions. To reach the optimal level of deploying SDN technology in enterprise environments, it is important to consider re-evaluating current deployed security solutions in traditional networks before deploying them to SDN-based infrastructures. Although DDoS attacks are a bit sinister, there are other types of cyber-threats that are very harmful, sophisticated, and intelligent. Thus, current security defense solutions to detect DDoS cannot detect them. These kinds of attacks are complex, persistent, and stealthy, also referred to Advanced Persistent Threats (APTs) which often leverage the bot control and remotely access valuable information. APT uses multiple stages to break into a network. APT is a sort of unseen, continuous and long-term penetrative network and attackers can bypass the existing security detection systems. It can modify and steal the sensitive data as well as specifically cause physical damage the target system. In this dissertation, two cyber-attack motivations are considered: sabotage, where the motive is the destruction; and information theft, where attackers aim to acquire invaluable information (customer info, business information, etc). I deal with two types of attacks (DDoS attacks and APT attacks) where DDoS attacks are classified under sabotage motivation category, and the APT attacks are classified under information theft motivation category. To detect and mitigate each of these attacks, I utilize the ease of programmability in SDN and its great platform for implementation, dynamic topology changes, decentralized network management, and ease of deploying security countermeasures.
ContributorsAlshamrani, Adel (Author) / Huang, Dijiang (Thesis advisor) / Doupe, Adam (Committee member) / Ahn, Gail-Joon (Committee member) / Davulcu, Hasan (Committee member) / Arizona State University (Publisher)
Created2018
Description
A story is defined as "an actor(s) taking action(s) that culminates in a resolution(s)''. I present novel sets of features to facilitate story detection among text via supervised classification and further reveal different forms within stories via unsupervised clustering. First, I investigate the utility of a new set of semantic features compared to standard keyword features combined with statistical features, such as density of part-of-speech (POS) tags and named entities, to develop a story classifier. The proposed semantic features are based on triplets that can be extracted using a shallow parser. Experimental results show that a model of memory-based semantic linguistic features alongside statistical features achieves better accuracy. Next, I further improve the performance of story detection with a novel algorithm which aggregates the triplets producing generalized concepts and relations. A major challenge in automated text analysis is that different words are used for related concepts. Analyzing text at the surface level would treat related concepts (i.e. actors, actions, targets, and victims) as different objects, potentially missing common narrative patterns. The algorithm clusters triplets into generalized concepts by utilizing syntactic criteria based on common contexts and semantic corpus-based statistical criteria based on "contextual synonyms''. Generalized concepts representation of text (1) overcomes surface level differences (which arise when different keywords are used for related concepts) without drift, (2) leads to a higher-level semantic network representation of related stories, and (3) when used as features, they yield a significant (36%) boost in performance for the story detection task. Finally, I implement co-clustering based on generalized concepts/relations to automatically detect story forms. Overlapping generalized concepts and relationships correspond to archetypes/targets and actions that characterize story forms. I perform co-clustering of stories using standard unigrams/bigrams and generalized concepts. I show that the residual error of factorization with concept-based features is significantly lower than the error with standard keyword-based features. I also present qualitative evaluations by a subject matter expert, which suggest that concept-based features yield more coherent, distinctive and interesting story forms compared to those produced by using standard keyword-based features.
ContributorsCeran, Saadet Betul (Author) / Davulcu, Hasan (Thesis advisor) / Corman, Steven R. (Committee member) / Shakarian, Paulo (Committee member) / Ye, Jieping (Committee member) / Arizona State University (Publisher)
Created2016
Description
Online health forums provide a convenient channel for patients, caregivers, and medical professionals to share their experience, support and encourage each other, and form health communities. The fast growing content in health forums provides a large repository for people to seek valuable information. A forum user can issue a keyword query to search health forums regarding to some specific questions, e.g., what treatments are effective for a disease symptom? A medical researcher can discover medical knowledge in a timely and large-scale fashion by automatically aggregating the latest evidences emerging in health forums.
This dissertation studies how to effectively discover information in health forums. Several challenges have been identified. First, the existing work relies on the syntactic information unit, such as a sentence, a post, or a thread, to bind different pieces of information in a forum. However, most of information discovery tasks should be based on the semantic information unit, a patient. For instance, given a keyword query that involves the relationship between a treatment and side effects, it is expected that the matched keywords refer to the same patient. In this work, patient-centered mining is proposed to mine patient semantic information units. In a patient information unit, the health information, such as diseases, symptoms, treatments, effects, and etc., is connected by the corresponding patient.
Second, the information published in health forums has varying degree of quality. Some information includes patient-reported personal health experience, while others can be hearsay. In this work, a context-aware experience extraction framework is proposed to mine patient-reported personal health experience, which can be used for evidence-based knowledge discovery or finding patients with similar experience.
At last, the proposed patient-centered and experience-aware mining framework is used to build a patient health information database for effectively discovering adverse drug reactions (ADRs) from health forums. ADRs have become a serious health problem and even a leading cause of death in the United States. Health forums provide valuable evidences in a large scale and in a timely fashion through the active participation of patients, caregivers, and doctors. Empirical evaluation shows the effectiveness of the proposed approach.
This dissertation studies how to effectively discover information in health forums. Several challenges have been identified. First, the existing work relies on the syntactic information unit, such as a sentence, a post, or a thread, to bind different pieces of information in a forum. However, most of information discovery tasks should be based on the semantic information unit, a patient. For instance, given a keyword query that involves the relationship between a treatment and side effects, it is expected that the matched keywords refer to the same patient. In this work, patient-centered mining is proposed to mine patient semantic information units. In a patient information unit, the health information, such as diseases, symptoms, treatments, effects, and etc., is connected by the corresponding patient.
Second, the information published in health forums has varying degree of quality. Some information includes patient-reported personal health experience, while others can be hearsay. In this work, a context-aware experience extraction framework is proposed to mine patient-reported personal health experience, which can be used for evidence-based knowledge discovery or finding patients with similar experience.
At last, the proposed patient-centered and experience-aware mining framework is used to build a patient health information database for effectively discovering adverse drug reactions (ADRs) from health forums. ADRs have become a serious health problem and even a leading cause of death in the United States. Health forums provide valuable evidences in a large scale and in a timely fashion through the active participation of patients, caregivers, and doctors. Empirical evaluation shows the effectiveness of the proposed approach.
ContributorsLiu, Yunzhong (Author) / Chen, Yi (Thesis advisor) / Liu, Huan (Thesis advisor) / Li, Baoxin (Committee member) / Davulcu, Hasan (Committee member) / Arizona State University (Publisher)
Created2016
Description
The dawn of Internet of Things (IoT) has opened the opportunity for mainstream adoption of machine learning analytics. However, most research in machine learning has focused on discovery of new algorithms or fine-tuning the performance of existing algorithms. Little exists on the process of taking an algorithm from the lab-environment into the real-world, culminating in sustained value. Real-world applications are typically characterized by dynamic non-stationary systems with requirements around feasibility, stability and maintainability. Not much has been done to establish standards around the unique analytics demands of real-world scenarios.
This research explores the problem of the why so few of the published algorithms enter production and furthermore, fewer end up generating sustained value. The dissertation proposes a ‘Design for Deployment’ (DFD) framework to successfully build machine learning analytics so they can be deployed to generate sustained value. The framework emphasizes and elaborates the often neglected but immensely important latter steps of an analytics process: ‘Evaluation’ and ‘Deployment’. A representative evaluation framework is proposed that incorporates the temporal-shifts and dynamism of real-world scenarios. Additionally, the recommended infrastructure allows analytics projects to pivot rapidly when a particular venture does not materialize. Deployment needs and apprehensions of the industry are identified and gaps addressed through a 4-step process for sustainable deployment. Lastly, the need for analytics as a functional area (like finance and IT) is identified to maximize the return on machine-learning deployment.
The framework and process is demonstrated in semiconductor manufacturing – it is highly complex process involving hundreds of optical, electrical, chemical, mechanical, thermal, electrochemical and software processes which makes it a highly dynamic non-stationary system. Due to the 24/7 uptime requirements in manufacturing, high-reliability and fail-safe are a must. Moreover, the ever growing volumes mean that the system must be highly scalable. Lastly, due to the high cost of change, sustained value proposition is a must for any proposed changes. Hence the context is ideal to explore the issues involved. The enterprise use-cases are used to demonstrate the robustness of the framework in addressing challenges encountered in the end-to-end process of productizing machine learning analytics in dynamic read-world scenarios.
This research explores the problem of the why so few of the published algorithms enter production and furthermore, fewer end up generating sustained value. The dissertation proposes a ‘Design for Deployment’ (DFD) framework to successfully build machine learning analytics so they can be deployed to generate sustained value. The framework emphasizes and elaborates the often neglected but immensely important latter steps of an analytics process: ‘Evaluation’ and ‘Deployment’. A representative evaluation framework is proposed that incorporates the temporal-shifts and dynamism of real-world scenarios. Additionally, the recommended infrastructure allows analytics projects to pivot rapidly when a particular venture does not materialize. Deployment needs and apprehensions of the industry are identified and gaps addressed through a 4-step process for sustainable deployment. Lastly, the need for analytics as a functional area (like finance and IT) is identified to maximize the return on machine-learning deployment.
The framework and process is demonstrated in semiconductor manufacturing – it is highly complex process involving hundreds of optical, electrical, chemical, mechanical, thermal, electrochemical and software processes which makes it a highly dynamic non-stationary system. Due to the 24/7 uptime requirements in manufacturing, high-reliability and fail-safe are a must. Moreover, the ever growing volumes mean that the system must be highly scalable. Lastly, due to the high cost of change, sustained value proposition is a must for any proposed changes. Hence the context is ideal to explore the issues involved. The enterprise use-cases are used to demonstrate the robustness of the framework in addressing challenges encountered in the end-to-end process of productizing machine learning analytics in dynamic read-world scenarios.
ContributorsShahapurkar, Som (Author) / Liu, Huan (Thesis advisor) / Davulcu, Hasan (Committee member) / Ameresh, Ashish (Committee member) / He, Jingrui (Committee member) / Tuv, Eugene (Committee member) / Arizona State University (Publisher)
Created2016