analysis

Documentation

Overview

Package analysis provides measurements and analysis tools for queries.

Index

• Variables
• func ConceptMatchReachability(queries []pipeline.Query, protocols guru.Protocols, conceptsBinFile string, ...) (conceptReachabilityMapping map[string][]ConceptReachability, ...)
• func ContainsMeshField(kw cqr.Keyword) bool
• func ExplodedKeywords(r cqr.CommonQueryRepresentation) (exploded []cqr.Keyword)
• func KeywordsWithField(r cqr.CommonQueryRepresentation, f string) (k []cqr.Keyword)
• func MeshExplodedKeywords(r cqr.CommonQueryRepresentation) (exploded []cqr.Keyword)
• func MeshNonExplodedKeywords(r cqr.CommonQueryRepresentation) (exploded []cqr.Keyword)
• func MetaMapScoreDistributions(queries []pipeline.Query, protocols guru.Protocols, client metawrap.HTTPClient) (inQuery map[string][]float64, notInQuery map[string][]float64, err error)
• func QueryBooleanClauseCount(r cqr.CommonQueryRepresentation, operator string) (n int)
• func QueryBooleanClauses(r cqr.CommonQueryRepresentation) (children []cqr.BooleanQuery)
• func QueryBooleanQueries(r cqr.CommonQueryRepresentation) (children []cqr.BooleanQuery)
• func QueryFields(r cqr.CommonQueryRepresentation) (fields []string)
• func QueryFieldsOfField(r cqr.CommonQueryRepresentation, field string) (fields int)
• func QueryKeywords(r cqr.CommonQueryRepresentation) (keywords []cqr.Keyword)
• func QueryTerms(r cqr.CommonQueryRepresentation) (terms []string)
• func TruncatedKeywords(r cqr.CommonQueryRepresentation) (truncated []cqr.Keyword)
• type ConceptReachability
• type DocumentOverlap
  • func NewDocumentOverlapMeasurement(document []string) DocumentOverlap
  • func (d DocumentOverlap) Execute(q pipeline.Query, s stats.StatisticsSource) (float64, error)
  • func (DocumentOverlap) Name() string
• type KeywordReachability
  • func StringMatchReachability(queries []pipeline.Query, protocols guru.Protocols) []KeywordReachability
• type Measurement
• type MeasurementCacher
• type MeasurementExecutor
  • func NewDiskMeasurementExecutor(d *diskv.Diskv) MeasurementExecutor
  • func NewMemoryMeasurementExecutor() MeasurementExecutor
  • func (m MeasurementExecutor) Execute(query pipeline.Query, ss stats.StatisticsSource, measurements ...Measurement) ([]float64, error)
• type MemoryMeasurementCache
  • func (m MemoryMeasurementCache) Read(key string) ([]byte, error)
  • func (m MemoryMeasurementCache) Write(key string, val []byte) error
• type QueryComplexity
  • func (qc QueryComplexity) Execute(q cqr.CommonQueryRepresentation, s stats.StatisticsSource) (float64, error)
  • func (qc QueryComplexity) Name() string

Variables

var (
	BooleanFields           booleanFields
	BooleanKeywords         booleanKeywords
	BooleanClauses          booleanClauses
	BooleanNonAtomicClauses booleanNonAtomicClauses
	BooleanTruncated        booleanTruncated
	BooleanFieldsTitle      booleanFieldsTitle
	BooleanFieldsAbstract   booleanFieldsAbstract
	BooleanFieldsMeSH       booleanFieldsMeSH
	BooleanFieldsOther      booleanFieldsOther
	BooleanAndCount         booleanAndCount
	BooleanOrCount          booleanOrCount
	BooleanNotCount         booleanNotCount
)

var (
	MeshKeywordCount     = meshKeywordCount{}
	MeshExplodedCount    = meshExplodedCount{}
	MeshNonExplodedCount = meshNonExplodedCount{}
	MeshAvgDepth         = meshAvgDepth{}
	MeshMaxDepth         = meshMaxDepth{}
)

var MeSHTree, _ = meshexp.Default()

var (
	// TermCount is a measurement that counts the number of terms in the query.
	TermCount = termCount{}
)

Functions

func ConceptMatchReachability

func ConceptMatchReachability(queries []pipeline.Query, protocols guru.Protocols, conceptsBinFile string, client metawrap.HTTPClient) (conceptReachabilityMapping map[string][]ConceptReachability, conceptsNotInTitle map[string]map[string]int, err error)

func ContainsMeshField

func ContainsMeshField(kw cqr.Keyword) bool

ContainsMeshField returns if a keyword query contains a mesh heading field.

func ExplodedKeywords

func ExplodedKeywords(r cqr.CommonQueryRepresentation) (exploded []cqr.Keyword)

ExplodedKeywords gets the keywords in the query that are exploded.

func KeywordsWithField

func KeywordsWithField(r cqr.CommonQueryRepresentation, f string) (k []cqr.Keyword)

MeshKeywords extracts all keywords which have a particular field.

func MeshExplodedKeywords

func MeshExplodedKeywords(r cqr.CommonQueryRepresentation) (exploded []cqr.Keyword)

MeshExplodedKeywords extracts all mesh heading keywords that are exploded.

func MeshNonExplodedKeywords

func MeshNonExplodedKeywords(r cqr.CommonQueryRepresentation) (exploded []cqr.Keyword)

MeshExplodedKeywords extracts all mesh heading keywords that are not exploded.

func MetaMapScoreDistributions

func MetaMapScoreDistributions(queries []pipeline.Query, protocols guru.Protocols, client metawrap.HTTPClient) (inQuery map[string][]float64, notInQuery map[string][]float64, err error)

func QueryBooleanClauseCount

func QueryBooleanClauseCount(r cqr.CommonQueryRepresentation, operator string) (n int)

QueryBooleanClauses extracts all of the clauses from a Boolean query, recursively.

func QueryBooleanClauses

func QueryBooleanClauses(r cqr.CommonQueryRepresentation) (children []cqr.BooleanQuery)

QueryBooleanClauses extracts all of the clauses from a Boolean query, recursively.

func QueryBooleanQueries

func QueryBooleanQueries(r cqr.CommonQueryRepresentation) (children []cqr.BooleanQuery)

QueryBooleanQueries extracts all of the sub-queries from a Boolean query, recursively.

func QueryFields

func QueryFields(r cqr.CommonQueryRepresentation) (fields []string)

QueryFields extracts the fields from a query.

func QueryFieldsOfField

func QueryFieldsOfField(r cqr.CommonQueryRepresentation, field string) (fields int)

QueryFields extracts the fields from a query.

func QueryKeywords

func QueryKeywords(r cqr.CommonQueryRepresentation) (keywords []cqr.Keyword)

QueryKeywords extracts the keywords from a query.

func QueryTerms

func QueryTerms(r cqr.CommonQueryRepresentation) (terms []string)

QueryTerms extracts the terms from a query.

func TruncatedKeywords

func TruncatedKeywords(r cqr.CommonQueryRepresentation) (truncated []cqr.Keyword)

TruncatedKeywords gets the keywords in the query that are exploded.

Types

type ConceptReachability

type ConceptReachability struct {
	QueryCount, FieldCount int
	Overlap, OverlapRatio  float64
	Topic                  string
}

type DocumentOverlap

type DocumentOverlap struct {
	// contains filtered or unexported fields
}

DocumentOverlap computes term overlap betweent the document and query.

func NewDocumentOverlapMeasurement

func NewDocumentOverlapMeasurement(document []string) DocumentOverlap

NewDocumentOverlapMeasurement creates an overlap measurement.

func (DocumentOverlap) Execute

func (d DocumentOverlap) Execute(q pipeline.Query, s stats.StatisticsSource) (float64, error)

Execute computes document overlap between query and document.

func (DocumentOverlap) Name

func (DocumentOverlap) Name() string

Name is the name for the document overlap measurement.

type KeywordReachability

type KeywordReachability struct {
	Title, Objectives, TypeOfStudy, Participants, IndexTests,
	TargetConditions, ReferenceStandards, Concepts int
	Topic string
}

func StringMatchReachability

func StringMatchReachability(queries []pipeline.Query, protocols guru.Protocols) []KeywordReachability

type Measurement

type Measurement interface {
	// Name is the name of the measurement in the output. It should not contain any spaces.
	Name() string
	// Execute computes the implemented measurement for a query and optionally using the specified statistics.
	Execute(q pipeline.Query, s stats.StatisticsSource) (float64, error)
}

Measurement is a representation for how a measurement fits into the pipeline.

type MeasurementCacher

type MeasurementCacher interface {
	Read(key string) ([]byte, error)
	Write(key string, val []byte) error
}

MeasurementCacher is a cache that can store the measurements for queries.

type MeasurementExecutor

type MeasurementExecutor struct {
	// contains filtered or unexported fields
}

MeasurementExecutor executes measurements while caching the results to improve performance.

func NewDiskMeasurementExecutor

func NewDiskMeasurementExecutor(d *diskv.Diskv) MeasurementExecutor

NewDiskMeasurementExecutor creates a measurement executor that caches to disk.

func NewMemoryMeasurementExecutor

func NewMemoryMeasurementExecutor() MeasurementExecutor

NewMemoryMeasurementExecutor creates a measurement executor that caches to memory.

func (MeasurementExecutor) Execute

func (m MeasurementExecutor) Execute(query pipeline.Query, ss stats.StatisticsSource, measurements ...Measurement) ([]float64, error)

Execute executes the specified measurements on the query using the statistics source.

type MemoryMeasurementCache

type MemoryMeasurementCache map[string][]byte

MemoryMeasurementCache caches measurement results in memory.

func (MemoryMeasurementCache) Read

func (m MemoryMeasurementCache) Read(key string) ([]byte, error)

Read reads a measurement result from memory.

func (MemoryMeasurementCache) Write

func (m MemoryMeasurementCache) Write(key string, val []byte) error

Write writes a measurement result to memory.

type QueryComplexity

type QueryComplexity struct{}

QueryComplexity is a measure of how "complex" a Boolean query is. It takes into account the number of keywords and Boolean queries in each of the top-level sub-contexts, with relation to the number of sub-contexts, and the size of the context (or query as a whole).

func (QueryComplexity) Execute

func (qc QueryComplexity) Execute(q cqr.CommonQueryRepresentation, s stats.StatisticsSource) (float64, error)

Execute computes query complexity.

func (QueryComplexity) Name

func (qc QueryComplexity) Name() string

Name is QueryComplexity