Michael E. Lesk

Computer Evaluation of Indexing and Text Processing

Automatic indexing methods are evaluated and design criteria for modern information systems are derived.

Relevance Assessments and Retrieval System Evaluation

Abstract Two widely used criteria for evaluating the effectiveness of information retrieval systems are, respectively, the recall and the precision. Since the determination of these measures is dependent on a distinction between documents which are relevant to a given query and documents which are not relevant to that query, it has sometimes been claimed that an accurate, generally valid evaluation cannot be based on recall and precision measures. A study was made to determine the effect of variations in relevance assessments on the average recall and precision values used to measure retrieval effectiveness. Using a collection of 1200 documents in information science for test purposes, it is found that large scale differences in the relevance assessments do not produce significant variations in average recall and precision. It thus appears that properly computed recall and precision data may represent effectiveness indicators which are generally valid for many distinct user classes.

Spectroscopic Constants and Vibrational Assignment for the B3Π0u + State of Iodine

Portions of the iodine B 3Π0u+←X 1Σg+ absorption spectrum have been remeasured under high resolution. A new analysis of 719 lines in the 3–6, 4–7, 5–4, 7–5, 11–1, 12–2, 13–2, 25–0, and 29–0 bands, combined with previous measurements by Mecke, Loomis, and Brown, gives the following revised constants: ωe′=125.273u2009cm−1Te′=15770.59γe′=−4.0×10−7ωexe′=0.7016Be′=0.028969δe′=−3.5×10−8ωeye′=−0.00567re′=3.0276u2009ADe′=3.5×10−9ωeze′=+0.000032αe′=0.0001562βe′=3.9×10−10De′=4391.0 Constants for ground state rotational levels were found to be in good agreement with Ranks latest determinations.Using these new constants for the B state, Franck—Condon factors for this system were computed. The vibrational numbering originally assigned by Mecke and Loomis to the B state was decreased by one unit to bring calculated intensity distribution into complete agreement with all observed fluorescence data, including new photoelectric measurements on a number of bands. No systematic variation of the electronic transition moment could b...

The SMART automatic document retrieval systems—an illustration

A fully automatic document retrieval system operating on the IBM 7094 is described. The system is characterized by the fact that several hundred different methods are available to analyze documents and search requests. This feature is used in the retrieval process by leaving the exact sequence of operations initially unspecified, and adapting the search strategy to the needs of individual users. The system is used not only to simulate an actual operating environment, but also to test the effectiveness of the various available processing methods. Results obtained so far seem to indicate that some combination of analysis procedures can in general be relied upon to retrieve the wanted information. A typical search request is used as an example in the present report to illustrate systems operations and evaluation procedures .

Fine Structure and Dipole Moment of Metastable a3Π Carbon Monoxide

The radio‐frequency spectrum of the first four (υu2009=u20090–3) vibrational states of the a 3Π metastable state of 12C16O has been measured in molecular beam electric resonance from 0 to 500 MHz. Lambda doubling transitions have been observed in Ju2009=u20092–7 of Ωu2009=u20092, and in Ju2009=u20091 and 2 of Ωu2009=u20091 at zero electric and magnetic field and as a function of electric field. An exhaustive reanalysis of a 3Π optical data is presented, in which the spin–orbit, rotational, spin–spin dipole–dipole, and potential‐energy parameters have been accurately determined. Together with parameters determined from the rf spectrum, the resulting set of fine structure parameters predicts the complete lambda doubling spectrum to better than 1 in 104 for Ju2009=u20090–50, υu2009=u20090–3. The predicted frequencies below 5 GHz are tabulated for possible radio‐astronomical measurement of a 3Π CO. The Stark effect allows the a 3Π dipole moment to be determined to an absolute accuracy of 2 in 104, and the small dependence upon spin–orbit, vibrational, and rotation...

Metastable a 3Π 13CO: Molecular‐Beam Electric‐Resonance Measurements of the Fine Structure, Hyperfine Structure, and Dipole Moment

The radio‐frequency spectrum of metastable a 3Π 13CO has been measured in molecular‐beam electric resonance in the range 0–2000 MHz. Electric‐dipole transitions with ΔJu2009=u20090 and ΔFu2009=u20090, ±u20091 have been observed in states |u2009υ, Ω, J, F〉u2009=u2009|u20090–3, 1–2, 1–7, Ju2009±u200912〉 at zero field. The set of transitions in each vibrational state has been analyzed to give three parameters describing the lowest‐order magnetic hyperfine interactions of the 13C nuclear spin (Iu2009=u200912) with the spin (S) and orbital (L) angular momenta of the two unpaired electrons in the 5σ2π open‐shell configuration. The vibrational dependence of the measured hyperfine parameters is fitted to a polynomial in (υu2009+u200912)n to give: OriginParameterValue at Re(a 3π) (in MHz)I·SFermi contactK=1935.1u2009±u20090.5I·Sspin–spin dipoleDδ=106.5u2009±u20090.1D=7.9u2009±u20090.1I·Lspin–orbitalG=162.5u2009±u20090.1 Restricted Hartree–Fock calculations of these hyperfine terms are in good agreement with the measured values. The Stark effect has been measured in states |u2009υ, Ω, J, F〉u2009=u2009|u20090–3, 2, 2–3, J...

Interactive search and retrieval methods using automatic information displays

Throughout the world, the design and operation of large-scale information systems has become of concern to an ever-increasing segment of the scientific and professional population. Furthermore, as the amount and complexity of the available information has continued to grow, the use of mechanized or partly mechanized procedures for various information storage and retrieval tasks has also become more widespread. While a number of retrieval systems are already in operation in which the search operations needed to compare the incoming information requests with the stored items are performed automatically, no systematic study has ever been made of the use of man-machine interaction as a part of a mechanized text analysis and information processing system. Specifically, the recent development of high capacity random-access storage mechanisms and conversational input-output consoles should permit a rapid interchange of information between users and system. Such an interchange can then be used to produce improved search formulations, resulting in a more effective retrieval service.

Making a digital library: the chemistry online retrieval experiment

The CORE project is an electronic library of primary journal articles in chemistry, containing about five years of twenty primary journals published by the American Chemical Society (about 425,000 pages). Unlike many digital library projects, CORE includes both a scanned image and a marked-up ASCII version (represented in Standard Generalized Markup Language, or SGML) for each page of the publishers database. Each page was scanned and segmented, with graphical units isolated and linked to figure references in the articles. The original machine-readable typography was converted to SGML format and the results were used to build databases with indexes for full-text Boolean searching; a single search engine served data for each of three X-Window interfaces.

Performance of automatic information systems

Abstract The SMART document retrieval system is used to investigate algorithms for text analysis and request searching. Results from three document collections indicate that word normalization is efficiently performed by automatic thesaurus lookup, while phrase matching procedures, statistical association methods, and concept hierarchies are useful for special applications. Automatic document clustering schemes and user-interactive feedback methods permit rapid searches of large collections. Abstracts are found to be superior to titles as a base for content analysis in a document retrieval system and almost as good as complete texts. Proper procedures for designing dictionaries and searching requests are discussed. The practicality of large scale document centers and their proper design are considered in light of these results.

The mindset of dependability

Computer software is legendary for its production time and cost overruns, and for its fragility after it is written. The U.S. government failed trying to procure dependable software for the IRS and the FAA, and the U.K. government was recently accused of wasting more than one billion pounds on failed or overdue information technology contracts. Perhaps only 25% of major software projects work out well. Home computer users are also accustomed to crashes. Why are computer systems so unreliable and difficult? By contrast, the Japanese Shinkansen trains are a remarkable testimony to reliability and safety. Since their inception in 1964, carrying millions of people per year, no passenger has been killed as a result of a collision, derailment, or other railway accident. Not only are the Shinkansen safe, they are also reliable. The average Shinkansen train arrives within 24 seconds of schedule. What can be learned from this? On one level, there are details of railway construction. The Shinkansen track is laid with heavier rail and closer-spaced cross-ties than a new line in Aus-tralia that will carry trains twice the weight. On another level, safety benefits from Japanese culture. Any visitor can tell you that Japan is an extremely clean country; the Shinkansen tracks and stations are litter-free. The worst fire ever on the London Underground (Kings Cross, 1987) started in debris under an escalator; cleanliness is not just cosmetic. But historically, Japan was not renowned for railway safety. As recently as the early 1960s, just before the Shinkansen opened, two accidents near Tokyo each killed more than 100 people. And yet safety has now become routine. The culture of safety and dependability has been learned there; it could be learned elsewhere. But Communications is neither a railway engineering journal nor a journal of cultural history. What should we learn about computers? The Japanese did not do a cost-benefit analysis on safety. Nobody sat in the Shinkansen design office and thought about how to trade off cutting construction costs against the number of people that would be killed. In the computer context, we often distribute the costs of unreliable software over a great many users who do not easily aggregate their frustrations into economic impact. NIST recently estimated that software bugs cost the U.S. economy