System Summary and Timing
  Organization Name: MDS at RMIT
  List of Run ID's: MDS001 MDS002 MDS003

  Construction of Indices, Knowledge Bases, and other Data Structures 

    Methods Used to build Data Structures 

    - Length (in words) of the stopword list: none 
    - Controlled Vocabulary?:  no 
    - Stemming Algorithm:   Lovins            
      - Morphological Analysis: NO 
    - Term Weighting:  IDF 
    -  Phrase Discovery?:  no            
    -  Syntactic Parsing?:  no 
    -  Word Sense Disambiguation?: no 
    -  Heuristic Associations (including short definition)?: no 
    -  Spelling Checking (with manual correction)?:  no 
    -  Spelling Correction?:  no 
    -  Proper Noun Identification Algorithm?:  no 
    -  Tokenizer?:              
    -  Manually-Indexed Terms?:  no 
    -  Other Techniques for building Data Structures:  no 

    Statistics on Data Structures built from TREC Text

    - Inverted index           
      - Run ID: MDS001 
      - Total Storage (in MB):  Not Applicable 
      - Total Computer Time to Build (in hours): Not Applicable  
      - Automatic Process? (If not, number of manual hours): Not Applicable
    -  Search Times             
      - Run ID:  MDS002
      - Computer Time to Search (Average per Query, in CPU seconds): Because 
awk scripts are used extensively to run many processes, we do not have CPU 
times available. Thus elapsed times are given: 75 sec elapsed time.
      - Component Times:  First search: 20 sec elapsed time
        Term expansion: 15 sec elapsed time
        Second search: 30 sec elapsed time
        Combination: 10 sec elapsed time (Estimates due to awk scripts.)
    -  Search Times             
      - Run ID: MDS003 
      - Computer Time to Search (Average per Query, in CPU seconds): 75 sec 
elapsed time
      - Component Times: First search: 20 sec elapsed time
        Term expansion: 15 sec elapsed time
        Second search: 30 sec elapsed time
        Combination: 10 sec elapsed time (Estimates due to awk scripts.)

    Factors in Ranking

      - Term Frequency?:  yes 
      - Inverse Document Frequency?:  yes 
      - Other Term Weights?:  no
      - Semantic Closeness?: no 
      - Position in Document?: no 
      - Syntactic Clues?: no 
      - Proximity of Terms?: no 
      - Information Theoretic Weights?: no 
      - Document Length?:  yes, pivoted 
      - Percentage of Query Terms which match?:  no 
      - N-gram Frequency?: no 
      - Word Specificity?:  no 
      - Word Sense Frequency?:  no 
      - Cluster Distance?: no  
      - Other: no  

    Machine Information

    - Machine Type for TREC Experiment:  Sparc 10 (4 cpus) 
    - Was the Machine Dedicated or Shared: shared 
    - Amount of Hard Disk Storage (in MB): 20,000 
    - Amount of RAM (in MB): 245 
    - Clock Rate of CPU (in MHz): 50 

    System Comparisons 

    - Given appropriate resources:            
      - Could your system run faster?: Yes. For MDS001, the system was 
simulated. Exhaustive search was used to compute similarities.  We are planning
to use the information learnt in these experiments as input to a new version of
mg.  For MDS002, 003 simple awk scripts were used for finding terms for 
expansion, and for combining. Both of these processes were much lower than 
need be.  
    - Features the System is Missing that would be beneficial:  Phrases