Thesis Topics

On this page you will find the list of topics for diploma and study theses. Applications and inquires should be sent to martin .welsch@de.ibm.com or koenig @informatik.uni-jena.de.

Topic 01: The Next Generation Contextual Portal (vergeben)

• Previously a lot of prototypes have been developed which now exist dispersed throughout different platforms and demo machines.
• As part of this thesis the student is responsible for migrating, merging and extending these solutions. He is responsible for architecting, designing and implementing a consolidated solution as well as for the design of a prototype suitable for demonstrating the benefits we gain. Of course, the student is given enough freedom to extend the existing solutions with his own ideas.

Topic 02: A System Architecture Supporting Context-Adaptive Web Portals (vergeben)

• Previously our solutions addressed single issues separately. They tried to improve chosen aspects such as the navigation, the layout of single pages etc.
• As part of this thesis the student is responsible for architecting, designing and implementing a concise system architecture on which all existing and future solutions can base. Components have to be developed allowing for user tracking, user modeling, context modeling and adaptation in general.

Topic 03: User Modeling for Context-Adaptive Web Portals (vergeben)

• The Contextual Portal heavily relies on accurate user models that reflect users’ behavior since these form the basis for any kind of adaptation. Ideally a system's responses should contain exactly the information that will be most helpful to the user. But since not all users are alike, achieving such behavior requires that the system has a model of the particular user with whom it is currently interacting This model should include information about the user's knowledge, beliefs, goals and plans for achieving these goals, abilities, attitudes, and preferences.
• Simple mechanisms to model users behavior already exist and have to be refined as part of this work. The student is responsible for evaluating reasonable metrics to track users’ behavior and for developing appropriate user models reflecting it.

Topic 04: Context Modeling for Context-Adaptive Web Portals

• Users’ behavior (cp. topic 03) varies over time. Changes in his context might influence his needs and tastes and result in changing requirements. A context-adaptive Web Portal should recognize the context a user is acting in and perform reasonable adaptations.
  In general, context is any information that can be used to characterize the situation of an entity. An entity is a person, place, or object that is considered relevant to the interaction between a user and an application, including the user and applications themselves. I.e. that context is any instantaneous, detectable, and relevant property of the environment, system, or users. One of the most important and widely used contexts is location e.g.
• The student is responsible for evaluating reasonable metrics to track the context users are acting in and for developing appropriate context models. Existing adaptation mechanisms have to be extended to not only allow for a general-, but also a context-sensitive kind of adaptation.

Topic 05a: (Machine) Learning in Context-Adaptive Web Portals

• As a broad subfield of artificial intelligence, machine learning, is concerned with the design and development of algorithms and techniques that allow computers to "learn". The major focus of machine learning research is to extract information from data automatically, by computational and statistical methods. Hence, machine learning is closely related to data mining and statistics but also theoretical computer science.
• We want to evaluate existing algorithms from within the field of machine learning w.r.t. their suitability for our project. The student is responsible for doing the necessary research, as well as for implementing prototypical components.

Topic 05b: Web (Usage) Mining for Context-Adaptive Web Portals

• Web mining is the application of data mining techniques to discover usage patterns within web technology in order to better understand and serve the needs of users of web-based applications. It is an activity that involves the automatic discovery of patterns. Web mining underlies the assumption that the available data is sufficiently structured to apply algorithms in order to search for patterns.
  Today web mining can generally be divided into three major areas: Web content mining focuses on techniques for searching the web for documents whose content meets users' queries. Web structure mining could be used to discover authority sites for the subjects (authorities) and overview site for the subjects that point to many authorities (hubs).
  Web usage mining can be viewed as the extraction of usage patterns from access log data containing the behavior characteristics of users.
• Within the scope of the Contextual Portal project web mining techniques might provide powerful mechanism to understand users’ behavior. The student is responsible for doing the necessary research, as well as for implementing prototypical components.

Topic 06: Analyzing User Behavior in (Enterprise Information) Web Portals (vergeben)

• Understanding users’ behavior is essential to design a proper context-adaptive Web Portal solution. Understanding behavior means to understand when, how often, in which order he uses which part of the system.
• As part of this work the student is responsible for statistically analyzing users’ behavior within Web Portal environments.

Topic 07: Clustering Users in Context-Adaptive Web Portals (vergeben)

• Previously, our solutions analyzed single users’ behavior only. Of course, usually sets of users can exist that behave similar. If we were able to recognize users behaving similarly we can analyze them as one entity and perform adaptations for one user part of this entity based on what other users part of this entity already did.
• As part of this thesis the student is responsible for finding means to cluster groups of users behaving similar. He is also responsible for extending (architecting, designing and implementing) existing solutions to make them analyzing entire groups.

Topic 08: Context-Adaptive Navigation in Web Portals (vergeben)

• The presentation of the navigation topology is a key factor in every Portal solution, which determines to some degree the Portals overall usability. Since the amount of time a user spends navigating to the target content (or navigation time) is considered valueless from his point of view, it is important to reduce it as much as possible and to increase the amount of time the users spend viewing or using the content (or content time) instead.
• As part of this work the student has to extend existing means for navigating in Web Portals.
• Interesting topics include: Voronoi Treemaps, Jigsaw Maps etc.

Topic 09: Context-Adaptive Page Layouts in Web Portals

• The presentation of pages layout is another key factor in every Portal solution, which determines to some degree the Portals overall usability. Preventing the overloading of pages by letting single portlets appear and disappear as needed is an interesting field.
• As part of this work the student has to extend existing means for presenting pages in Web Portals.

Topic 10: Context-Adaptive Portlets in Web Portals

• Today portlets are mainly installed by Portal administrators. Usually they are not preconfigured, as the administrator is not interested in using these portlets in some business context, nor does he have the knowledge/time to preconfigure each portlet he installed meaningfully (for every single Portal user (group)).
  This leads to scenarios where every Portal user has to configure his portlet himself, even if other Portal users within the community might already have done some similar configuration. Letting more than one user within a community perform similar configuration tasks of the portlets used, costs time (hence, money) due to the fact that the similar work is done redundantly. Besides that, there might be (new) users acting in the community which do not know how to configure a certain portlet accordingly. They which would highly appreciate a meaningful preconfiguration.
  Currently there is no way to exchange portlet configurations among users in a Portal community - besides having a phone call or exchange mails etc. to manually synchronize the settings. And, moreover, preconfigured portlets cannot be exchanged due to the fact that Portals utilize a centralized administration model and provide no way for decentralized exchange of portlet (data).
• As part of this work the student has to evaluate mechanisms allowing for storing (and loading) of portlet’s configuration and for providing channels to (decentralized) distribute preconfigured portlets or the sole configuration data among Portal users.
• Also interesting would be a comparison of how similar things have been realized with Google Gadgets and within iGoogle.

Topic 11: Exchanging Contextual Information in Web Portals

• Context-adaptive systems have to provide mechanisms that allow for tagging of context information for later retrieval. Within Web Portals pages and portlets are seldom used totally unrelated. A user that works with its Portal often uses a set of pages and portlets to accomplish its work. Often these pages and portlets are even used in a cooperative manner, i.e. together to accomplish the same single work. Hence, often data has to be exchanged between these pages and portlets.
• As part of this work the student is responsible for finding ways to exchange data across application boundaries.

Topic 12: Sensor Framework to React on Contextual Changes in Web Portals

• If the attributes that might influence the context a user is acting in changes he needs to be notified about this change and/or certain actions being triggered. So, e.g., if new, high priority, tasks are assigned to a user via a workflow system his current work might become less important – his context changes.
• As part of this work the student is responsible for architecting, designing and implementing a generic, extensible sensor framework that observe users environment. As said, sensors might observe tasks available in a workflow system and notify users and/or trigger appropriate actions.

Topic 13: A Feedbacking System to Detect User Context

• When blending in recommendations or performing adaptation we could also observe the users’ reaction to this kind of assistance. The main question is: Does he use the help we provide?
  This analysis might allow us to determine his current context and especially the point in time when it changes more accurately.
• As part of this thesis the student is responsible for doing the necessary research, as well as for implementing prototypical components.

Topic 14: Unstructured Data Analysis for Adaptation in Web Portals (vergeben)

• Unstructured information represents the largest, most current and fastest growing source of knowledge available to businesses and governments world-wide.
  Unstructured information includes the documents found on the web, plus an estimated 80% of the information generated by enterprises around the world. The principal challenge with unstructured information is that it needs to be analyzed in order to identify, locate and relate the entities and relationships of interest – discover the vital knowledge contained therein.
  Once these entitles and relationships are detected they may be indexed in structured forms so that powerful search technologies like search engines and database engines can efficiently find the knowledge you need, when you need it.
  Web Portals consist of unstructured information, too. Especially with the upcoming Web 2.0 Hype components like Wikis & Blogs create more and more unstructured information repositories.
• Finding means to analyze and “understand” this information allows for enhanced methods of adaptation in Web Portals. The student is responsible for integrating IBM’s UIMA framework in order to analyze unstructured information available within Web Portals in a way which allows for proper adaptation of content.

Topic 15a: On the Synergy of Annotations and Semantic Tags for Adaptation in Web Portals (vergeben)

• A tag is a (relevant) keyword or term associated with or assigned to a piece of information (e.g. a picture, article, or video clip), thus describing the item and enabling keyword-based classification of information. Hence, tagging is the process of assigning tags to resources.
  Within Web Portals a tag engine we have already developed allows any resource (page, portlet, user, …) part of the Portal to be tagged. This enables users to categorize content autonomously by themselves. A tag cloud (visual depiction of available tags) allows users to get an overview of tags already been used and can be used to navigate/search through available content; it provides them with quick access to categorized content (or even „expert users“).
  Semantic tagging allows certain content to be identified as a resource with certain semantics (e.g. that two words represent people’s names).
• As part of this work we try to allow resources being semantically tagged to be tagged via collaborative tagging, too. Analyzing options for the usage of machine tags and microformats might be interesting, too.

Topic 15b: Next Generation Tagging (vergeben)

• A tag is a (relevant) keyword or term associated with or assigned to a piece of information (e.g. a picture, article, or video clip), thus describing the item and enabling keyword-based classification of information. Hence, tagging is the process of assigning tags to resources.
  Within Web Portals a tag engine we have already developed allows any resource (page, portlet, user, …) part of the Portal to be tagged. This enables users to categorize content autonomously by themselves. A tag cloud (visual depiction of available tags) allows users to get an overview of tags already been used and can be used to navigate/search through available content; it provides them with quick access to categorized content (or even „expert users“).
  How can the next generation of tag engines look like?
• As part of this thesis the student is responsible for doing the necessary research, as well as for implementing prototypical components.
• Interesting topics include: Machine tagging/tags, hierarchical tagging, recommended tagging, adaptive tagging, synonym resolver for tagging engines, DnD tagging, aggregation of external tags etc.

Topic 16a: On the Use of Web 2.0 Methodologies for Adaptaton in Web Portals

• The Web 2.0 changes the way the Web is used and content created. Advocates of the concept suggest that technologies such as weblogs, social bookmarking, wikis, podcasts, RSS feeds (and other forms of many-to-many publishing), social software, Web APIs, Web standards and online Web services imply a significant change.
  Web 2.0 refers to the the transition of websites from isolated information silos to sources of content and functionality, thus becoming computing platforms serving web applications to end-users. It also is a social phenomenon embracing an approach to generating and distributing Web content itself, characterized by open communication, decentralization of authority, freedom to share and re-use, and "the market as a conversation" . It allows for the enhanced organization and categorization of content, emphasizing deep linking a rise in the economic value of the Web, possibly surpassing.
• As part of this work the student is responsible for evaluating options allowing for reasonable adaptation in a Web 2.0 world.
• Interesting topics include: analysis of blogs, tag clustering and visualization, synonym and homonym resolution in tags, visual and textual information extraction, temporal analysis, data streams, trend detection and concept drift, application of web and text mining to wiki content, discovering social structures and communities, evolution of online social networks, predicting user behavior, analysis of dynamic networks, discovering misuse and fraud, combining the web with data from other sources, mining with mashups, deriving profiles from usage, personalized delivery of information, applications, case studies etc.

Topic 16b: Social Computing Methodologies for Adaptation in Web Portals (vergeben)

• Social computing is a general term for an area of computer science that is concerned with the intersection of social behavior and computational systems. It has become more widely known because of its relationship to a number of recent trends. These include the growing popularity of social software and Web 2.0, increased academic interest in social network analysis, the rise of open source as a viable method of production, and a growing conviction that all of this can have a profound impact on daily life.
• Interesting topics deal with the implicit construction of social networks in Web Portals.
  Can we construct social networks, can we do social classification just by analyzing and comparing users behavior?

Topic 17a: Semantic Context-Adaptive Web Portals

• Semantic Portals allow for integrated and syndicated data views on information by using ontological knowledge and machine processable semantic descriptions. In particular, Semantic Portals should allow the user to customize the access to information by making advanced use of the semantic descriptions of the information, and should provide individualized views on the data, enhance user awareness and orientation, help the user in detecting relevant information or relations, etc.
• As part of this work the student is responsible for defining suitable ontologies describing Web Portal content as well as methods to perform reasoning for adaptation on the basis of the ontological knowledge.

Topic 17b: Ontology Based Adaptation in Web Portals

• Similar to topic 17a.

Topic 18a: Adaptive Mashups (vergeben)

• A mashup is a web application that combines data from more than one source into an integrated experience. Content used in mashups is typically sourced from third parties. Dynamic, context-adaptive construction of mashups without or with less user interaction represents an interesting field of research.
• As part of this thesis the student is responsible for doing the necessary research, as well as for implementing prototypical components.

Topic 18b: Automated Construction of Comp. Apps. Based on User Defined Constraints (vergeben)

• In computing, the term composite application expresses a perspective of software engineering that defines an application built by combining multiple services. A composite application consists of functionality drawn from several different sources within a service oriented architecture (SOA). The components may be individual web services, selected functions from within other applications, or entire systems whose outputs have been packaged as web services (often legacy systems).
  In Web Portals composite applications usually refer to a set of pages and portlets representing applications fulfilling a dedicated logic.
  Today these composite applications are created based on the static information in templates. Any modification to the application is instance specific and has to be done to all other applications manually again. Any modification is purely user driven the user has to find reasonable modifications for himself. I.e. that no hints are provided by the system.
  As part of this work the student has to architecture, design and implement a system allowing for modifications of composite applications based on common usage patterns. Therefore, modifications to applications and usage patterns within applications are tracked/recognized and partitions of users behaving "similar" identified. Based on the analysis of this information the user is provided with recommendations on how the application could be modified/used/customized.
  Any proposal can be scoped by the application role of the user (role specific patterns, filtering of the proposals for best efficiency, etc.).
  The proposals and adaptations can also depend on contextual properties (current date, time, location, etc.) or other constraints in order to respect the fact that user behavior is highly related to the context he is acting in.

Topic 19: Authoring Context-Adaptive Web Portals

• When designing context-adaptive Web Portals an interesting question is how administrators can define/influence the system’s behavior during its design time.
• As part of this thesis the student is responsible for doing the necessary research, as well as for implementing prototypical components.

Topic 20a: (Context-sensitive) Adaptation in Web Portals: Dos and Don’ts (vergeben)

• What do users like, what do they not like and why?
  How efficient are the single solutions that have been developed?
• As part of this thesis the student is responsible for doing a usability study based on the current implementation of the Contextual Portal.

Topic 20b: Psychology, Ergonomic and Usability Aspects within Context-Adaptive Web Portals (vergeben)

• Similar to topic 20a.

Topic 21: Spreadsheet-based data transformation for portlet events

• One of the great benefits of Portlets is that they can send / receive events of a defined payload type from / to other portlets. Many use cases will include lists of properties as event values. Sometimes the properties need to be transformed in order to comply with another portlets expected payload type. Typically such transformations would combine properties from different fields into one property (e.g. first name and last name into name) or split properties from one field into multiple fields (e.g. address into street, city, zip code, state, country). In addition some fields may require some computations to be applied in order to transforms the value (e.g. birth date to age). In order to perform these transformations spreadsheet-like technologies should be used.
• As part of this thesis the student will
  • integrate spreadsheet transformations as special wire into the portal infrastructure
  • create a mapping portlet, displaying two property lists and allows graphically to wire/combine/split properties
  • propagate the server model into a Web 2.0 model in order to allow such transformations for client side events too

Other topics

• Context-Adaptive Search in Web Portals
• Context-Adaptive Help Systems for Web Portals
• Automated Personalization and Rule Construction in Web Portals
• Content Clustering Based On User Behavior - A Recommendation Algebra
• On the Synergy of Activity Centric Computing and Context-Adaptive Web Portals
• Topic Trend Detection in Web Portals
• Spreading-activation based Adaptation in Web Portals
• ...