Indiana University-Purdue University Indianapolis (IUPUI) / Web search has become a part of everyday life for hundreds of millions of users around the world. However, the effectiveness of a user's search depends vitally on the quality of search result ranking. Even though enormous efforts have been made to improve the ranking quality, there is still significant misalignment between search engine ranking and an end user's preference order. This is evident from the fact that, for many search results on major search and e-commerce platforms, many users ignore the top ranked results and click on the lower ranked results. Nevertheless, finding a ranking that suits all the users is a difficult problem to solve as every user's need is different. So, an ideal ranking is the one which is preferred by the majority of the users. This emphasizes the need for an automated approach which improves the search engine ranking dynamically by incorporating user clicks in the ranking algorithm. In existing search result ranking methodologies, this direction has not been explored profoundly. A key challenge in using user clicks in search result ranking is that the relevance feedback that is learnt from click data is imperfect. This is due to the fact that a user is more likely to click a top ranked result than a lower ranked result, irrespective of the actual relevance of those results. This phenomenon is known as position bias which poses a major difficulty in obtaining an automated method for dynamic update of search rank orders. In my thesis, I propose a set of methodologies which incorporate user clicks for dynamic update of search rank orders. The updates are based on adaptive randomization of results using reinforcement learning strategy by considering the user click activities as reinforcement signal. Beginning at any rank order of the search results, the proposed methodologies guaranty to converge to a ranking which is close to the ideal rank order. Besides, the usage of reinforcement learning strategy enables the proposed methods to overcome the position bias phenomenon. To measure the effectiveness of the proposed method, I perform experiments considering a simplified user behavior model which I call color ball abstraction model. I evaluate the quality of the proposed methodologies using standard information retrieval metrics like Precision at n (P@n), Kendall tau rank correlation, Discounted Cumulative Gain (DCG) and Normalized Discounted Cumulative Gain (NDCG). The experiment results clearly demonstrate the success of the proposed methodologies.

Indiana University-Purdue University Indianapolis (IUPUI) / While research links the presence of meaning in life to better psychological well-being, the relationship between the search for meaning and psychological well-being is less clear. The search for meaning is generally thought to be psychologically distressing, but there is evidence that this process is moderated by the presence of meaning in life. Because the search for meaning in life can be considered a goal pursuit, goal-related personality traits may also moderate the relationship between the search for meaning and psychological well-being. The first aim of this cross-sectional study was to replicate the moderating effect of the presence of meaning on the relationship between the search for meaning and psychological well-being in a sample of undergraduates (N = 246). The second aim was to examine the potential moderating effects of hope and optimism on these relationships. As an exploratory third aim, this study examined whether there was a unique combination of the presence of meaning, the search for meaning, and hope or optimism that differentially predicted psychological well-being. Results suggest that optimism and the presence of meaning, but not hope, are significant moderators of the relationship between the search for meaning and psychological well-being. Implications and limitations of these findings are discussed.

No description available.

Research into the nature and content of visual long-term memory has investigated what aspects of its representation may account for the remarkable ability we have to remember large amounts of detailed visual information. One theory proposed is that visual memories are supported by an underlying structure of conceptual knowledge around which visual information is organized. However, findings in memory for visual information learned in a visual search task were not explained by this theory of conceptual support, and a new theory is proposed that incorporates the importance of episodic, task-relevant visual information into the organizational structure of visual memory. The current study examined visual long-term memory organization as evidenced by retroactive interference effects in memory for objects learned in a visual search. Four experiments were conducted to examine the amount of retroactive interference induced based on aspects in which interfering objects were related to learned objects. Specifically, episodically task-relevant information about objects was manipulated between conditions based on search instructions. Aspects of conceptual category, perceptual information (color), and context (object role in search) were examined for their contribution to retroactive interference for learned objects. Findings indicated that when made episodically task-relevant, perceptual, as well as conceptual, information contributed to the organization of visual long-term memory. However, when made episodically non-relevant, perceptual information did not contribute to memory organization, and memory defaulted to conceptual category organization. This finding supports the theory of an episodically defined organizational structure in visual long-term memory that is overlaid upon an underlying conceptual structure.

Bayesian networks are a widely used graphical model which formalize reasoning under uncertainty. Unfortunately, construction of a Bayesian network by an expert is timeconsuming, and, in some cases, all expertsmay not agree on the best structure for a problem domain. Additionally, for some complex systems such as those present in molecular biology, experts with an understanding of the entire domain and how individual components interact may not exist. In these cases, we must learn the network structure from available data. This dissertation focuses on score-based structure learning. In this context, a scoring function is used to measure the goodness of fit of a structure to data. The goal is to find the structure which optimizes the scoring function. The first contribution of this dissertation is a shortest-path finding perspective for the problem of learning optimal Bayesian network structures. This perspective builds on earlier dynamic programming strategies, but, as we show, offers much more flexibility. Second, we develop a set of data structures to improve the efficiency of many of the integral calculations for structure learning. Most of these data structures benefit our algorithms, dynamic programming and other formulations of the structure learning problem. Next, we introduce a suite of algorithms that leverage the new data structures and shortest-path finding perspective for structure learning. These algorithms take advantage of a number of new heuristic functions to ignore provably sub-optimal parts of the search space. They also exploit regularities in the search that previous approaches could not. All of the algorithms we present have their own advantages. Some minimize work in a provable sense; others use external memory such as hard disk to scale to datasets with more variables. Several of the algorithms quickly find solutions and improve them as long as they are given more resources. Our algorithms improve the state of the art in structure learning by running faster, using less memory and incorporating other desirable characteristics, such as anytime behavior. We also pose unanswered questions to drive research into the future.

The United States has undergone a dramatic demographic shift in the last 40 years, particularly in terms of the racial and ethnic composition of the country. Like the other racial and ethnic minority groups, the Hispanic population is also a rapidly increasing segment of the United States. These shifts have affected many of the country's social institutions. For example, primary and secondary education have experienced a dramatic demographic shift in terms of race and ethnicity in the last 20 years. Higher education has also been affected by demographic shifts in the U.S. Although the numbers of racial and ethnic minorities enrolled in college have increased, that growth has not been proportionate to the changing numbers in the United States' population. For example, in 1990 the census reported over 22 million Hispanics in the U.S., or 9% of the total population, yet the 724,600 Hispanic students enrolled in higher education represented only 6% of all students in college ("College Enrollment,", 1998; "We the Americanâ ¦Hispanics", 1993). These gaps between Hispanic growth in the general population and Hispanic college enrollment are due to several factors, including the college choice process. Currently, research on Hispanic college choice is limited. The purpose of this study was to examine the factors that influence the college choice process for Hispanic students. Data were collected by administering the College Choice Survey (CCS), an instrument designed specifically for this study. A total of 383 surveys were mailed and 144 surveys were completed and returned by respondents. This reflected a response rate of 38%. A total of 65 ANOVAs were run on the data elicited from participants. Five ANOVAs were run on the subscales, which included total College Choice Survey scores, Internal Search scores, Internal Selection scores, External Search scores, and External Selection scores. The dependent variables were gender, generational status, and ethnic background. A total of three significant differences were found among these five ANOVAs. The remaining 60 ANOVAs examined differences reported by respondents on the last two items in the survey. These items asked participants to rate the degree to which they used sources of support for both the search and selection processes. The ANOVAs were run for differences by main effect only (i.e. gender, generational status, and ethnic background). Results revealed a total of three significant differences on the sources of support participants used during the search process and a total of four significant differences among the sources of support respondents used during the selection process. In summary, this study was valuable because it contributed to the understanding of the college choice process of Hispanic students. The results of this study revealed both pragmatic and significant differences in the college choice process of Hispanic students by gender, generational status, and ethnic background. Higher education administrators may strive to better understand the differences in the college choice process of Hispanic students and consider these differences in designing recruitment and admissions efforts. / Master of Arts

No description available.

We conduct a parallel Hunter search in order to find a degree 7 number field K ramified at primes q and p with discriminant d(K)=q^6 p^2 where q=11 and 2

School officials trying to deter drug use, combat crime, and shore up security are conducting searches that are landing school in legal trouble for violating students' constitutional rights. In 1993, West Virginia Supreme Court ruled that a strip search of a student suspected of stealing money was illegal (State of West Virginia ex rel Gilford v. Mark Anthony B., 1993). In another case, a federal appellate court held that a strip search of a student for suspected drug possession was reasonable, although no drugs were found (Cornfield v. Consolidated High School District No. 230, 1993). Improper searches of students, lockers and automobiles can result in hundreds of thousands of dollars in civil liability, costs and attorney fees. This study collected data on administrative knowledge in the area of search and seizure. The analyzed data served to (a) determine if administrators across the state of Florida have a general understanding of the laws regarding search and seizure; (b) identify demographic areas that demonstrate a lack of knowledge related to search and seizure; and (c) suggest improvements to current educational leadership courses of study, state-wide staff development offerings, and ideas for possible conference topics. The study involved responses from questionnaires received from 139 public school administrators in Florida (17% of the 810 randomly sampled elementary, middle, and high school principals). Analysis of data revealed that more than one-third of the respondents fell below the mean, with no significant difference between building levels or metropolitan statistical area.

The main purpose of this thesis is to develop a search strategy for space debris that are in the geosynchronous orbit (GEO) region. The search strategy is not an effort to find the object initially but rather if found one time to aid in finding it again within a small time frame. This was a request from NASA Johnson Space Center Orbital Debris Program Office through the MODEST, Michigan Orbital Debris Survey Telescope, program. A single definitive search pattern was not found, however depending on the COEs of the orbit specific search strategy can be employed. These search strategies are far from perfect and can be improved upon with more rigorous testing as well as a larger data sample. Another goal is to look for correlation between the orbital parameters and the errors in the predicted right ascension (RA) and the declination (DEC). This was accomplished by varying the different orbital parameters by ±10% individually while holding the other parameters constant. This showed some correlation existed between some parameters and their errors, in particular there was correlation between a variation in right ascension of ascending node (RAAN) and the value of RAAN itself. The correlation found was that with the higher the value of RAAN the larger the RMS error.