《《IR&WMLes00-2016S》.pdf

InformationInformation RetrievalRetrieval && WebWeb MiningMining Course Introduction Dong Shoubin (董守斌) sbdong@ Communication & Computer Network Laboratory (CCNL) ,SCUT （（华南理工大学广东省计算机网络重点实验室华南理工大学广东省计算机网络重点实验室）） Information Retrieval (IR) & Web Mining • Information retrieval: the retrieval of documents Concerned firstly with retrieving relevant documents to a query. Concerned secondly with retrieving from large sets off documents effiffi ciientlly . • WebWeb miningmining Discovery, analysis and utilization of useful informationinformation fromfrom thethe WorldWorld WideWide WebWeb 2 “We are living in exponential times “ Information Overload ((信息过载信息过载)) • “I’m definingg information overload as a state of having more information available that one can readilyy assimilate,, that is,, ppeopple have difficultyy absorbing the information into their base of knowledgge. This hinders decision-makingg and judgment by causing stress and cognitive imppediments such as confusion,, uncertaintyy and distraction” - SteveSteve BellerBeller Why Information Overload? • A rapidly increasing rate of new information being produced • The ease of duplication and transmission of data across thethe InternetInternet • An increase in the available channels of incoming information ((e.gg. telepphone,, e-mail,, instant messagging,g, RSS) • Large amounts of historical information to dig through • Contradictions and inaccuracies in available information • AA llow siignall-tto-noiise ratitio • A lack of a method for comparing and processing differentdifferent kindskinds ofof informationinformation How to Deal With i