Recommender system for real estate management/Rekomendacine nekilnojamojo turto valdymo sistema.
Ginevicius, Tomas ; Kaklauskas, Arturas ; Kazokaitis, Paulius 等
1. Recommender systems and recommendation methods
Definitions of recommender systems can be found in different
literature sources as follows:
--Any system that produces individualized recommendations as output
or has the effect of guiding the user in a personalized way to
interesting or useful objects in a large space of possible options
(Burke 2002).
--Recommender system provides users with a ranked list of the
recommended items (Herlocker et al. 2004).
--[...] people provide recommendations as inputs, which the system
then aggregates and directs to appropriate recipients (Resnick, Varian
1997).
--Recommender systems form a specific type of information filtering
(IF) technique that attempts to present information items that are
likely of interest to the user. Typically, a recommender system compares
the user's profile to some reference characteristics, and seeks to
predict the 'rating' that a user would give to an item they
had not yet considered. These characteristics may be from the
information item (the content-based approach) or the user's social
environment (Recommender systems 2011).
Adomavicius and Tuzhilin (2005) present an overview of the field of
recommender systems and describe the current generation of
recommendation methods that are usually classified into the following
three main categories: content-based, collaborative, and hybrid
recommendation approaches. Adomavicius and Tuzhilin (2005) also describe
various limitations of current recommendation methods and discuss
possible extensions that can improve recommendation capabilities and
make recommender systems applicable to an even broader range of
applications. These extensions include, among others, an improvement of
understanding of users and items, incorporation of the contextual
information into the recommendation process, support for multi-criteria
ratings, and a provision of more flexible and less intrusive types of
recommendations.
Recommender systems have been evaluated in many, often
incomparable, ways. Herlocker et al. (2004) review the key decisions in
evaluating collaborative filtering recommender systems: the user tasks
being evaluated, the types of analysis and datasets being used, the ways
in which prediction quality is measured, the evaluation of prediction
attributes other than quality, and the user-based evaluation of the
system as a whole. In addition to reviewing the evaluation strategies
used by prior researchers, Herlocker et al. (2004) present empirical
results from the analysis of various accuracy metrics on one content
domain where all the tested metrics collapsed roughly into three
equivalence classes. Metrics within each equivalency class were strongly
correlated, while metrics from different equivalency classes were
uncorrelated.
Recommender systems represent user preferences for the purpose of
suggesting items to purchase or examine. They have become fundamental
applications in electronic commerce and information access, providing
suggestions that effectively prune large information spaces so that
users are directed toward those items that best meet their needs and
preferences. A variety of techniques have been proposed for performing
recommendation, including content-based, collaborative, knowledge-based
and other techniques. To improve performance, these methods have
sometimes been combined in hybrid recommenders (Burke 2002). Burke
(2002) surveys the landscape of actual and possible hybrid recommenders,
and introduces a novel hybrid, EntreeC, a system that combines
knowledge-based recommendation and collaborative filtering to recommend
restaurants. Burke (2002) shows that semantic ratings obtained from the
knowledge-based part of the system enhance the effectiveness of
collaborative filtering.
Sarwar et al. (2000) investigate the use of dimensionality
reduction to improve performance for a new class of data analysis
software called "recommender systems". Recommender systems
apply knowledge discovery techniques to the problem of making product
recommendations during a live customer interaction. These systems are
achieving widespread success in E-commerce nowadays, especially with the
advent of the Internet. The tremendous growth of customers and products
poses three key challenges for recommender systems in the E-commerce
domain. These are: producing high quality recommendations, performing
many recommendations per second for millions of customers and products,
and achieving high coverage in the face of data sparseness. One
successful recommender system technology is collaborative filtering,
which works by matching customer preferences to other customers in
making recommendations. Collaborative filtering has been shown to
produce high quality recommendations, but the performance degrades with
the number of customers and products. New recommender system
technologies are needed that can quickly produce high quality
recommendations, even for very large-scale problems (Sarwar et al.
2000).
When building the user's profile a distinction is made between
explicit and implicit forms of data collection. Examples of explicit
data collection include the following: asking a user to rate an item on
a sliding scale; asking a user to rank a collection of items from
favorite to least favorite; presenting two items to a user and asking
him/her to choose the best one; asking a user to create a list of items
that he/she likes. Examples of implicit data collection include the
following: observing the items that a user views in an online store;
analyzing item/user viewing times; keeping a record of the items that a
user purchases online; obtaining a list of items that a user has
listened to or watched on his/her computer; analyzing the user's
social network and discovering similar likes and dislikes. The
recommender system compares the collected data to similar data collected
from others and calculates a list of recommended items for the user.
Several commercial and non-commercial examples are listed in the article
on collaborative filtering systems. Recommender systems are a useful
alternative to search algorithms since those help users discover items
they might not have found by themselves. Interestingly enough,
recommender systems are often implemented using search engines indexing
non-traditional data. (Recommender systems 2011). Adomavicius and
Tuzhilin (2005) provide an overview of recommender systems. Herlocker et
al. (2004) provide an overview of evaluation techniques for recommender
systems.
One of the most commonly used algorithms in recommender systems is
Nearest Neighborhood approach (Sarwar et al. 2000). In a social network,
a particular user's neighborhood with similar taste or interest can
be found by calculating Pearson Correlation, by collecting the
preference data of top N nearest neighbors of the particular user
(weighted by similarity), the user's preference can be predicted by
calculating the data using certain techniques. Another family of
algorithms that is widely used in recommender systems is Collaborative
Filtering. One of the most common types of Collaborative Filtering is
item-to-item collaborative filtering (people who buy x also buy y), an
algorithm popularized by Amazon.com's recommender system
(Recommender systems 2011). The Netflix Prize, a contest with a dataset
of over 100 million movie ratings and a grand prize of $1,000,000, has
energized the search for new and more accurate algorithms. The most
accurate algorithm in 2007 used 107 different algorithmic approaches,
blended into a single prediction. Predictive accuracy is substantially
improved when blending multiple predictors. Our experience is that most
efforts should be concentrated in deriving substantially different
approaches, rather than refining a single technique. Consequently, our
solution is an ensemble of many methods (Bell et al. 2007).
2. Determination of rational real estate management processes
A real estate management process consists of closely interrelated
stages: i.e. consultation, planning, procurement, implementation and
monitoring. A real estate management process may have many alternative
versions. These variants are based on alternative consultation,
planning, procurement, implementation and controlling stages and their
constituent parts. The above solutions and processes will be considered
in more detail later. For instance, alternative space management
variants can be developed by varying their space organization, removals,
inventory compilation/updating and main services solutions (building
security, reception, telephone switchboard, cleaning, snow-clearing
service, upkeep of outdoor real estate, garden care, plant care in the
building, post room, travel office, office service, central secretariat,
canteen management, removals service, central archive, courier services,
office supplies and safety specialist). Therefore, thousands of real
estate management process alternative versions can be obtained. The
diversity of available solutions contributes to a more accurate
evaluation of climatic conditions, risk exposure, maintenance services,
as well as making the project cheaper and results in a better way of
satisfying a client's aesthetic, comfort, technological and other
requirements. This also leads to a better satisfaction of the needs of
all the involved parties in the real estate management process.
Various interested parties (e.g. users, owners, and real estate
managers, etc.) are involved in the real estate management process, and
trying to satisfy their needs and affecting its efficiency.
The above needs or objectives include the expected cost of real
estate management services, occupier, owner and building support,
building inspection, budgeting, cost optimization, coordination of
services, accounting. It also includes contract management; leasing
management; technical operations management; maintenance, inspection
and, repair of equipment and systems (gas, water, wastewater, heating,
ventilation, cooling, electrical systems, lifts, warehousing systems,
automatic door and gate, communication, cable and network, security,
laundry and dry-cleaning systems, general building equipment, other
equipment and systems), etc. Real estate management companies should be
able to offer a range of services that can be flexibly extended or
reduced.
The problem is how to define an efficient real estate management
process when many various parties are involved because the alternative
versions come to thousands and the efficiency changes with the
alterations in the business environment conditions and the constituent
parts of the process. Moreover, the realization of some objectives seems
more rational from the economic perspective though from other
perspectives (i.e. technological, comfort, space, administrative,
technical, etc.) they have various significances. Therefore, it is
considered that the efficiency of a real estate management process
depends on the rationality of its stages as well as on the ability to
satisfy the needs of the interested parties and the rational character
of micro and macro-level environment conditions.
A formalized presentation of the research shows how changes in the
business environment and the extent to which the goals pursued by
various interested parties are satisfied, cause corresponding changes in
the value and utility degree of the real estate management process. With
this in mind, it is possible to solve the problem of optimization
concerning the satisfaction of the needs with reasonable expenditures.
This requires an analysis of the real estate management process versions
allowing one to find an optimal combination of pursued goals and
available finances.
The determination of the utility degree and value of the real
estate management alternatives under investigation and the establishment
of the priority order for their implementation do not present much
difficulty if the criteria numerical values and weights have been
obtained and the multiple criteria decision making methods are used.
By way of an illustration, we provide a short analysis of a
criteria system of some real estate management constituent parts. They
include computer-aided real estate management systems, service of a real
estate, and equipment.
Cormier (2000) described the process and elements for comparison
and the selection when considering various computeraided real estate
management systems. Cormier (2000) considered the following criteria
system: modules and tools (lease management, move management, strategic
space planning, maintenance management, accounting/charge-back,
communication/cable management, personnel management, etc.) and also
considered cost information (cost of software, cost of training, cost
and ease of software integration, cost of software maintenance, and
after-sale support), technical information (platform, network access,
native database support, database connectivity, user interface,
security, reports, file formats, and interoperability) and key features.
The service of a real estate can be evaluated as: operational
productivity, aesthetic value or public image, comfort (noise, colour,
air quality, thermal comfort, and working conditions) flexibility, and
cost (design, construction, indirect expenses, operating and maintenance
expenses, renovation costs, the interest paid on loan).
Effectiveness of equipment can be evaluated by the following
criteria system:
--Price;
--Expenses for use;
--Expenses for repair (maintenance, capital);
--Capacity;
--Number of operations performed;
--Reliability;
--Comfort;
--Physical and technical durability;
--Weight.
One of the main tasks of the efficient implementation of real
estate management is multiple criteria optimization of its life cycle
process with the aim of maximum purpose satisfaction of all interested
parties in the process. The interested parties and their aspired goals
make up one entity. However, there are some potential conflicts among
interested parties: e.g. speed versus waste, cost versus quality,
capital cost now versus after-operational efficiency, aesthetics and
comfort versus cost, environment versus user needs, etc. The greater the
scope of the realization of pursued goals (taking into account their
significance) the greater (in the opinion of interested parties) the
total efficiency of the project. In other words, the total efficiency of
a project is directly proportional to the entity of its realized goals.
3. Multiple criteria analysis of real estate management
alternatives
Most of all calculators, analysers, software, neural networks,
decision support and expert systems seek to find out how to make the
most economic real estate management decisions and most of all these
decisions are intended only for economic objectives. Real estate
management alternatives under evaluation have to be evaluated not only
from an economic position, but take into consideration qualitative,
technical, technological and other characteristics as well. For example,
an analysis of the service of real estate is usually performed by taking
into account operational productivity, aesthetic value or public image,
comfort (noise, colour, air quality, thermal comfort, working
conditions), flexibility, and cost (design, construction, indirect
expenses, operating and maintenance expenses, renovation costs, the
interest paid on loan). Real estate management alternative solutions
allow for a more rational and realistic assessment of economic,
technical, technological, space conditions and traditions and for
greater satisfaction of different customer requirements. Therefore, by
applying multiple criteria analysis methods and decision support systems
the efficiency of real estate management calculators, analysers,
software, neural networks, decision support and expert systems may be
increased.
Bauer et al. (2000) discussed six major real estate phases which
include the following: definition of need, planning and programming,
design, construction, operate/ maintenance and decision for use the next
time. According to Bauer et al. (2000) each of these phases has five
process groups called: initiating, planning, execution, controlling and
closing. On that score, a real estate management's life cycle has
many alternative versions. Variants are based on the project's
alternatives of the definition of need, planning and programming,
design, construction, operate/maintenance and other processes. The above
solutions and processes may be further considered in more detail. For
example, there are several ways that companies can provide necessary
cleaning services (Smith et al. 2000): in a traditional department, all
personnel are company employees; in support of a traditional department,
some companies are adding their services to a competent consultant; the
company can use a management service to support its own production team;
in a full-service program, a service company provides all the management
and production personnel, tools, equipment and supplies; in combination
programs, the company uses its employees to perform part of the cleaning
responsibilities and the contracts with a service company for the
remainder.
Thousands of real estate management's life cycle alternative
versions can be obtained in this way.
The determination of the utility degree and market value of the
real estate management alternatives under investigation and the
establishment of the priority order for their implementation do not
present much difficulty if the criteria numerical values and weights
have been obtained and the multiple criteria decision-making methods are
used.
4. A method of multiple criteria complex proportional evaluation
and defining the utility degree of real estate
This method is directly related with utilitarianism moral
philosophy. For example, for Bentham (1948), conduct is to be judged by
its consequences to the community. Actions are moral to the extent that
they promote the community's utility, and immoral to the extent
that these lessen it. Utility is understood in subjective terms as the
net balance of whatever a person finds to be pleasurable and painful,
with the former obviously increasing that balance and the latter
decreasing it. Rather than being conceived holistically as an entity in
its own right, the community is nothing more than the name we give to a
collection of individuals. Accordingly, Bentham holds that the
community's utility is the sum of individual utilities. It can be
calculated by placing the number of those positively impacted by an
action, weighted by the intensity and duration of their net pleasure, in
the positive column and then doing the same in the negative column for
those negatively affected by net pain. If the positive side of the
ledger exceeds the negative, communal utility rises and the action
passes the moral bar, and vice versa, if the negative column outweighs
the positive column (Bentham 1948).
The determination of the utility degree and value of the
alternative under investigation and establishment of the priority order
for their implementation do not present much difficulty if the criteria
numerical values and weights have been obtained and the multiple
criteria decision making methods are used.
All criteria are calculated for the whole alternative. The process
of determining the system of criteria, their initial weights and
qualitative criteria numerical values of the alternative under
investigation is based on the use of various expert methods. The
determination of quantitative criteria numerical values is based on the
use of various statistical methods, analysed alternatives,
recommendations and other documents.
The results of the comparative analysis of the alternatives are
presented as a grouped decision making matrix where columns contain n
alternatives being considered, while all quantitative and conceptual
information pertaining to them is found in lines.
Quantitative and conceptual description of the research object
provides the information about various aspects of a real estate life
cycle (i.e. economic, legal/regulatory, institutional, political,
social, traditions, cultural, philosophical, ethical, confidence,
happiness, religion, emotional, psychological, etc.). Quantitative
information is based on the criteria systems and subsystems, units of
measure, values and initial weights as well as the data on the
alternatives development.
Conceptual description of a real estate life cycle presents
textual, graphical (schemes, graphs, diagrams, drawings), visual
(videotapes) information about the alternatives and the criteria used
for their definition, as well as giving the reason for the choice of
this particular system of criteria, their values and weights. This part
also includes information about the possible ways of multi-variant
design. Conceptual information is needed to make more complete and
accurate evaluation of the alternatives considered. It also helps to get
more useful information as well as developing a system and subsystems of
criteria and defining their values and weights.
In order to perform a complete study of the research object, a
complex evaluation of its economic, legal/regulatory, institutional,
social, traditions, cultural, philosophical, ethical, confidence,
happiness, religion, emotional, psychological and other aspects is
needed. The diversity of aspects being assessed should follow the
diversity of ways of presenting data needed for decision making.
Therefore, the necessary data may be presented in numerical, textual,
graphical (schemes, graphs, charts), formula, videotape and other forms.
The grouping of the information in the matrix should be performed
so as to facilitate the calculation process and to express their
physical meaning. In our case the criteria system is formed from the
criteria describing the life cycle of real estate which can be expressed
in a quantitative form (quantitative criteria) and the criteria
describing the life cycle of real estate which cannot be expressed in a
quantitative form (qualitative criteria).
The researchers from various countries engaged in the analysis of
real estate life cycle and its stages did not consider the research
object being project by the authors of the present investigation.
However, they did not consider the research object that the research
project presented here (see section "Conceptual Model of Real
estate in Lithuania"). This research object may be described as a
life cycle of the real estate that includes the stakeholders involved
and the environment which impact a life cycle in some particular manner,
thus forming an integral, whole entity. This formulated research object
underwent complex analysis performed with the help of the multiple
criteria analysis, a new method specially developed for this purpose: a
method of multiple criteria complex proportional evaluation and defining
the utility degree of real estate.
This method assumes direct and proportional dependence of
significance and priority of investigated versions on a system of
criteria adequately describing the alternatives and on values and
weights of the criteria. The system of criteria is determined and the
values and initial weights of criteria are calculated by experts. All
this information can be corrected by interested parties taking into
consideration their pursued goals and existing capabilities. Hence, the
assessment results of alternatives fully reflect the initial data
jointly submitted by experts and interested parties.
The determination of significance and priority of alternatives is
carried out in four stages.
Stage 1. The weighted normalized decision making matrix D is
formed. The purpose of this stage is to receive dimensionless weighted
values from the comparative indexes. When the dimensionless values of
the indexes are known, all criteria, originally having different
dimensions, can be compared. The following formula is used for this
purpose:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (1)
where [x.sub.ij]--the value of the i-th criterion in the j-th
alternative of a solution; m--the number of criteria; n--the number of
the alternatives compared; [q.sub.i]--significance of i -th criterion.
The sum of dimensionless weighted index values [d.sub.ij] of each
criterion xi is always equal to the significance [q.sub.i] of this
criterion:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (2)
In other words, the value of significance qi of the investigated
criterion is proportionally distributed among all alternative versions
[a.sub.j] according to their values [x.sub.ij].
Stage 2. The sums of weighted normalized indexes describing the
j-th version are calculated. The versions are described by minimizing
indexes [S.sub.-j] and maximizing indexes [S.sub.+j]. The lower value of
minimizing indexes is better (price of the plot and real estate, etc.).
The greater value of maximizing indexes is better (comfortability and
aesthetics of the real estate, etc.). The sums are calculated according
to the formula:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (3)
In this case, the values S+j (the greater is this value
(alternative 'pluses'), the more satisfied are the interested
parties) and [S.sub.-j] (the lower is this value (alternative
'minuses'), the better is goal attainment by the interested
parties) express the degree of goals attained by the interested parties
in each alternative. In any case the sums of 'pluses'
[S.sub.+j] and 'minuses' [S.sub.-j] of all alternatives are
always respectively equal to all sums of significances of maximizing and
minimizing criteria:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (4)
In this way, the calculations made may be additionally checked.
Stage 3. The significance (efficiency) of comparative versions is
determined on the basis of describing positive alternatives
('pluses') and negative alternatives ('minuses')
characteristics. Relative significance [Q.sub.j] of each alternative
[a.sub.j] is found according to the formula:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (5)
Stage 4. Priority determination of alternatives. The greater is the
Qj the higher is the efficiency (priority) of the alternative.
The analysis of the method presented makes it possible to state
that it may be easily applied to evaluating the alternatives and
selecting most efficient of them, being fully aware of a physical
meaning of the process. Moreover, it allowed to formulate a reduced
criterion [Q.sub.j] which is directly proportional to the relative
effect of the compared criteria values [x.sub.ij] and significances
[q.sub.i] on the end result.
Significance [Q.sub.j] of real estate [a.sub.j] indicates
satisfaction degree of demands and goals pursued by the interested
parties--the greater is the [Q.sub.j] the higher is the efficiency of
the real estate. In this case, the significance [Q.sub.max] of the most
rational real estate will always be the highest. The significances of
all remaining real estate are lower as compared with the most rational
one. This means that total demands and goals of interested parties will
be satisfied to a smaller extent than it would be in case of the best
real estate.
The degree of real estate utility is directly associated with
quantitative and conceptual information related to it. If one real
estate is characterized by the best economic and political aspects,
while the other shows better social, philosophical and ethical
characteristics, both having obtained the same significance values as a
result of multiple criteria evaluation, this means that their utility
degree is also the same. With the increase (decrease) of the
significance of real estate analyzed, its degree of utility also
increases (decreases). The degree of real estate utility is determined
by comparing the real estate analysed with the most efficient real
estate. In this case, all the utility degree values related to the real
estate analysed will be ranged from 0% to 100%. This will facilitate
visual assessment of real estate efficiency.
The degrees of utility of the real estate considered as well as the
market value of the real estate being valuated are determined in seven
stages.
Stage 1. The formula used for the calculation of real estate
[a.sub.j] utility degree [N.sub.j] is given below:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (6)
here [Q.sub.j] and [Q.sub.max] are the significances of the real
estate obtained from the equation 5.
The degree of utility [N.sub.j] of real estate [a.sub.j] indicates
the level of satisfying the needs of the parties interested in the real
estate. The more goals are achieved and the more important they are, the
higher is the degree of the real estate utility. Since stakeholders are
mostly interested in how much more efficient particular real estate are
than the others (which can better satisfy their needs), then it is more
advisable to use the concept of real estate utility rather than
significance when choosing the most efficient solution.
A degree of real estate utility reflects the extent to which the
goals pursued by the interested parties are attained. The more
objectives are attained and the more significant they are, the higher
will be real estate degree of utility.
5. Recommender System for Real Estate Management
Based on the analysis of the existing information, expert and
decision support systems and in order to determine the most efficient
versions of real estate management a Recommender System for Real Estate
Management (RSREM) consisting of a database, database management system,
model-base, model-base management system and user interface was
developed.
5.1. Database
Real estate management involves a number of interested parties who
pursue various goals and have different potentialities, educational
levels and experiences. This leads to various approaches of the above
parties to decision-making in this field. In order to do a full analysis
of the available alternatives and to obtain an efficient compromise
solution, it is often necessary to analyse economic, qualitative, legal,
social, technical, technological and other types of information. This
information should be provided in a user-oriented way.
The presentation of information needed for decisionmaking in the
RSREM may be in a conceptual form (i.e. digital/numerical, textual,
graphical, diagrams, graphs and drawing, etc.), photographic, sound,
video and quantitative forms. The presentation of quantitative
information involves criteria systems and subsystems, units of
measurement, values and initial weights that fully define the provided
variants. Conceptual information means a conceptual description of the
alternative solutions, the criteria and ways of determining their values
and the weights, etc.
In this way, the RSREM enables the decision-maker to receive
various conceptual and quantitative information on real estate
management from a database and a model-base allowing him/her to analyse
the above factors and to form an efficient solution.
The analysis of database structures in decision support systems
according to the type of problem solved reveals their various utilities.
There are three basic types of database structures: hierarchical,
network and relational. RSREM has a relational database structure where
the information is stored in the form of tables. These tables contain
quantitative and conceptual information. Each table is given a name and
is saved in the computer's external memory as a separate file.
Logically linked parts of the table form a relational model.
The following tables form the RSREM's database:
--Initial data tables. These contain information about the real
estate (i.e. building and complexes).
--Tables assessing real estate management solutions. These contain
quantitative and conceptual information about alternative real estate
management solutions: market, competitors, suppliers, contractors,
renovation of walls, windows, roof, etc.
To design the structure of a database and perform its completion,
storage, editing, navigation, searching and browsing, etc., a database
management system was used in this research.
The user seeking for an efficient real estate management solution
should provide, in the tables assessing real estate management
solutions, the exact information about alternatives under consideration
as to the client's financial situation. It should be noted that
various users making a multiple criteria analysis of the same
alternatives often get diverse results. This may be due to the diversity
of the overall aims and financial positions of the users. Therefore, the
initial data provided by various users for calculating the real estate
management project differ and consequently lead to various final
results.
The character of the objective's choice for the most efficient
variant is largely dependent on all available information. It should
also be noted that the quantitative information is objective. The actual
real estate management services have real costs. The values of the
qualitative criteria are usually rather subjective though the
application of expert's methods contributes to their objectivity.
The interested parties have their specific needs and financial
situation. Therefore, every time when the party uses the RSREM they may
make corrections to the database according to their aims and their
financial situation. For example, a certain client considers the sound
insulation of the external walls to be more important than their
appearance while another client is quite of the opposite opinion. The
client striving to express his/her attitude towards these issues
numerically may ascribe various weights values to them that eventually
will affect the general estimation of a refurbishment project. Though
this assessment may seem biased and even quite subjective, the solution
finally made may exactly meet the client's requirements, aims and
affordability.
The tables assessing real estate management solutions are used as a
basis for working out the matrices of decisionmaking. These matrices,
along with the use of a model-base and models, make it possible to
perform a multiple criteria analysis of alternative real estate
management projects, resulting in the selection of the most beneficial
variants.
5.2. Model-base
The efficiency of a real estate management variant is often
determined by taking into account many factors. These factors include an
account of the economic, aesthetic, technical, technological, comfort,
legal, social and other factors. The model-base of a decision support
system should include models that enable a decision-maker to do a
comprehensive analysis of the available variants and to make a proper
choice. The following models of a model-base aim at performing the
functions of:
--a model for the establishment of the criteria weights,
--a model for multiple criteria analysis and for setting the
priorities,
--a model for the determination of a project's utility degree,
--a model for the determination of a project's market value,
--a model for the recommendations.
According to the user's needs, various models may be provided
by a model management system. When a certain model (i.e. search for real
estate management alternatives) is used the results obtained become the
initial data for some other models (i.e. a model for multiple criteria
analysis and setting the priorities). The results of the latter, in
turn, may be taken as the initial data for some other models (i.e.
determination of utility degree, market, suppliers, contractors,
renovation of walls, windows, roof, etc.).
The management system of the model base allows a person to modify
the available models, eliminate those that are no longer needed and add
some new models that are linked to the existing ones.
Since the analysis of real estate management is usually performed
by taking into account economic, quality, technical, technological,
legal, social and other factors, a modelbase should include models which
will enable a decisionmaker to carry out a comprehensive analysis of the
available variants and make a proper choice. The following multiple
criteria analysis methods and models as developed by the authors
(Zavadskas et al. 1994) are used by the RSREM in the analysis of the
real estate management alternatives:
1. A new method and model of complex determination of the weight of
the criteria taking into account their quantitative and qualitative
characteristics was developed. This method allows one to calculate and
co-ordinate the weights of the quantitative and qualitative criteria
according to the above characteristics.
2. A new method and model of multiple criteria complex proportional
evaluation of projects enabling the user to obtain a reduced criterion
determining the complex (overall) efficiency of the project was
suggested. This generalized criterion is directly proportional to the
relative effect of the values and weights of the considered criteria on
the efficiency of the project.
3. In order to find what price will make a valuated project
competitive on the market, a method and model for determining the
utility degree and market value of projects based on the complex
analysis of all their benefits and drawbacks was suggested. According to
this method the project's utility degree and the market value of a
project being estimated are directly proportional to the system of the
criteria and adequately describe them, the values and weights of these
criteria.
4. A new method and model of multiple criteria multi-variant design
of a project's life cycle enabling the user to make computer-aided
design of up to 100,000 alternative project versions was developed. Any
project's life cycle variant obtained in this way is based on
quantitative and conceptual information.
Application of Recommender System for Real Estate Management
(RSREM) allows one to determine the strengths and weaknesses of each
phase and its constituent parts. Calculations were made to find out by
what degree one version is better than another and the reasons disclosed
why it is namely so. Landmarks are set for an increase in the efficiency
of real estate management versions. All this was done argumentatively,
basing oneself on criteria under investigation and on their values and
weights. This saved users' time considerably by allowing them to
increase both the efficiency and quality of real estate management
analysis.
There is a list of typical real estate management problems that
were solved by users:
--Analysis of interested parties (competitors, suppliers,
contractors, etc.);
--Determination of efficient loans;
--Analysis and selection of rational refurbishment versions (e.g.
roof, walls, windows, etc.);
--Multiple criteria analysis and determination of the market value
of real estate (e.g. residential houses, commercial, office,
warehousing, manufacturing and agricultural buildings, etc.);
--Analysis and selection of a rational market;
--Determination of efficient investment versions, etc.;
--Providing recommendations.
6. Conclusions
Real estate management is an information business. Technological
innovation mainly through changes in the availability of information and
communication technology include calculators, analysers, software,
neural networks, decision support and expert systems that have been
provided by a variety of new services developed by the real estate
management sector. Most of all calculators, analysers, software,
decision support and expert systems, neural networks seek to find out
how to make the most economic real estate management decisions, and most
of all these decisions are intended only for economic objectives. Real
estate management alternatives under evaluation have to be evaluated not
only from the economic position, but take into consideration
qualitative, technical, technological and other characteristics as well.
Therefore, applying multiple criteria analysis methods and recommender
systems may increase the efficiency of real estate management
calculators, analysers, software, neural networks, decision support and
expert systems. Based on an analysis of existing information, expert and
decision support systems and in order to determine the most efficient
versions of real estate management, Recommender System for Real Estate
Management was developed by authors of the paper. The related questions
were also analysed in this paper.
doi: 10.3846/btp.2011.26
Received 21 February 2011; accepted 4 April 2011
Iteikta 2011-02-21; priimta 2011-04-04
References
Adomavicius, G.; Tuzhilin, A. 2005. Toward the Next Generation of
Recommender Systems: A survey of the state-of-the-art and possible
extensions, IEEE Transactions on Knowledge and Data Engineering 17(6):
734-749. doi:10.1109/TKDE.2005.99
Bauer, B. D.; Kenneth, H.; Stowe, K. H. 2000. Building life
cycle-an integrated approach, in Facilities Engineering and Management
Handbook. Commercial, Industrial, and Institutional Buildings.
McGraw-Hill, 87-93.
Bell, R.; Koren, Y.; Volinsky, C. 2007. The BellKor solution to the
Netflix Prize [cited 12 January 2011]. Available from Internet:
<http://www.netflixprize.com/assets/ProgressPrize2007_KorBell.pdf>.
Bentham, J. 1948. The Principles of Morals and Legislation. New
York: Hafner Press. 352 p.
Burke, R. D. 2002. Hybrid Recommender Systems, Survey and
Experiment, User Modeling and User-Adapted Interaction 12(4): 331-370.
doi:10.1023/A:1021240730564
Cormier, D. R. 2000. Comparison of FM Automation Systems.
Facilities Engineering and Management Handbook. Commercial, Industrial,
and Institutional Buildings. McGrawHill. 225 p.
Herlocker, J. L; Konstan, J. A.; Terveen, L. G; Riedl, J. 2004.
Evaluating Collaborative Filtering Recommender Systems, ACM Transactions
on Information Systems (TOIS) 22(1): 5-53. doi:10.1145/963770.963772
Recommender systems [cited 12 January 2011]. Available from
Internet: <http://en.wikipedia.org/wiki/Recommender_system>.
Resnick, P.; Varian, H. R. 1997. Recommender Systems,
Communications of the ACM 40(3): 56-58. doi:10.1145/245108.245121
Sarwar, B.; Karypis, G.; Konstan, J.; Riedl, J. 2000. Application
of Dimensionality Reduction in Recommender System A Case Study [cited 12
January 2011]. Available from Internet:
<http://glaros.dtc.umn.edu/gkhome/node/122>.
Smith, P. R.; Seth, A. K.; Wessel, R., et al. 2000. Facilities
Engineering and Management Handbook. Commercial, Industrial, and
Institutional Buildings. McGraw-Hill, 1200 p.
Zavadskas, E.; Peldschus, F.; Kaklauskas, A. 1994. Multiple
Criteria Evaluation of Projects in Construction. Institute of
Technological and Economic Development (ITED). Vilnius Technical
University: Technika. 226 p.
Tomas Ginevicius (1), Arturas Kaklauskas (2), Paulius Kazokaitis
(3), Jurgita Alchimoviene (4)
Vilnius Gediminas Technical University, Sauletekio al. 11, LT-10223
Vilnius, Lithuania
E-mails: (1) tg@oh.lt (corresponding author); (2)
Arturas.Kaklauskas@vgtu.lt; (3) paulius200@yahoo.com; (4)
jurgita.alchimoviene@vgtu.lt
Vilnius Gedimino technical universitetas, Sauletekio al. 11,
LT-10223 Vilnius, Lituva
El. pastas: (1) tg@oh.lt; (2) Arturas.Kaklauskas@vgtu.lt; (3)
paulius200@yahoo.com; (4) jurgita.alchimoviene@vgtu.lt
Tomas GINEVICIUS. PhD student of Vilnius Gediminas Technical
University, Department of Social Economics and Business Management.
Research interests: Project management, knowledge management, real
estate management.
Arturas KAKLAUSKAS. Professor of Vilnius Gediminas Technical
University, Department of Construction Economics and Property
Management. Research interests: Project management, knowledge
management, real estate management, multi-criteria evaluation, decision
support and recommender systems, etc.
Paulius KAZOKAITIS. PhD student of Vilnius Gediminas Technical
University, Department of Construction Economics and Property
Management. Research interests: Project management, knowledge
management, real estate management.
Jurgita ALChIMOVIENE. PhD student of Vilnius Gediminas Technical
University, Department of Construction Technology and Management,
Research interests: renovation and modernization of buildings, real
estate management.
Table 1. Grouped decision making matrix of real estate life cycle
multiple criteria analysis
Quantitative information pertinent to alternatives
Criteria describing
the life cycle * Weight Measuring
of a real estate units
[z.sub.1] [q.sub.1] [m.sub.1]
Quantitative [z.sub.2] [q.sub.2] [m.sub.2]
criteria ... ... ...
[z.sub.i] [q.sub.i] [m.sub.i]
[z.sub.t] [q.sub.t] [m.sub.t]
[z.sub.t+1] [q.sub.t+1] [m.sub.t+1]
Qualitative [z.sub.t+2] [q.sub.t+2] [m.sub.t+2]
criteria
[z.sub.i] [q.sub.i] [m.sub.i]
... ... ...
[z.sub.m] [q.sub.m] [m.sub.m]
Conceptual information pertinent to alternatives (i.e. text, drawings,
graphics, video tapes)
[C.sub.f] [C.sub.z] [C.sub.q] [C.sub.m]
Criteria describing
the life cycle * Compared alternatives
of a real estate [a.sub.1] [a.sub.2] ...
[z.sub.1] [x.sub.11] [x.sub.12] ...
Quantitative [z.sub.2] [x.sub.21] [x.sub.22]
criteria ... ... ... ...
[z.sub.i] [x.sub.i1] [x.sub.i2] ...
[z.sub.t] [x.sub.t1] [x.sub.t2] ...
[z.sub.t+1] [x.sub.t+1 1] [x.sub.t+1 2] ...
Qualitative [z.sub.t+2] [x.sub.t+2 1] [x.sub.t+2 2] ...
criteria
[z.sub.i] [x.sub.i1] [x.sub.i2] ...
... ... ...
[z.sub.m] [x.sub.m1] [x.sub.m2] ...
Conceptual information pertinent to alternatives (i.e. text, drawings,
graphics, video tapes)
[C.sub.z] [C.sub.1] [C.sub.2]
Criteria describing
the life cycle *
of a real estate [a.sub.j] ... s.n]
[z.sub.1] [x.sub.1j] ... [x.sub.1n]
Quantitative [z.sub.2] [x.sub.2j] ... [x.sub.2n]
criteri
a ... ... ...
[z.sub.i] [x.sub.ij] ... [x.sub.in]
[z.sub.t] [x.sub.tj] ... [x.sub.tn]
[z.sub.t+1] [x.sub.t+1 j] ... [x.sub.t+1 n]
Qualitative [z.sub.t+2] [x.sub.t+2 j] ... [x.sub.t+2 n]
criteria
[z.sub.i] [x.sub.ij] ... [x.sub.in]
... ... ...
[z.sub.m] [x.sub.mj] ... [x.sub.mn]
Conceptual information pertinent to alternatives (i.e. text, drawings,
graphics, video tapes)
[C.sub.z] [C.sub.j] [C.sub.n]
* The sign [z.sub.i] (+ (-)) indicates that a greater (less) criterion
value corresponds to a higher significance for stakeholders
Table 2. Real estate life cycle multiple criteria analysis results
Quantitative information pertinent to alternatives
Criteria describing * Weight Measuring Compared
the life cycle units alternatives
of a real estate
[a.sub.1]
[X.sub.1] [z.sub.1] [q.sub.1] [m.sub.1] [d.sub.11]
[X.sub.2] [z.sub.2] [q.sub.2] [m.sub.2] [d.sub.21]
[X.sub.3] [z.sub.3] [q.sub.3] [m.sub.3] [d.sub.31]
... ... ... ... ...
[X.sub.i] [z.sub.i] [q.sub.i] [m.sub.i] [d.sub.i1]
... ... ... ... ...
[X.sub.m] [z.sub.m] [q.sub.m] [m.sub.m] [d.sub.m1]
The sums of weighted [S.sub.+1]
normalized maximizing
(alternatives 'pluses')
indices of the alternative
The sums of weighted [S.sub.-1]
normalized minimizing
(alternatives 'minuses')
indices of the alternative
Significance of the [Q.sub.1]
alternative
Priority of the [P.sub.1]
alternative
Utility degree of the [N.sub.1]
alternative (%)
Criteria describing
the life cycle
of a real estate Compared alternatives
[a.sub.2] ... [a.sub.j] ... [a.sub.n]
[X.sub.1] [a.sub.12] ... [d.sub.1j] ... [d.sub.1n]
[X.sub.2] [d.sub.22] ... [d.sub.2j] ... [d.sub.2n]
[X.sub.3] [d.sub.32] ... [d.sub.3j] ... [d.sub.3n]
... ... ... ... ...
[X.sub.i] [d.sub.i2] ... [d.sub.ij] ... [d.sub.in]
... ... ... ... ... ...
[X.sub.m] [d.sub.m2] ... [d.sub.mj] ... [d.sub.mn]
The sums of weighted [S.sub.+2] ... [S.sub.+j] ... [d.sub.in]
normalized maximizing
(alternatives 'pluses')
indices of the alternative
The sums of weighted [S.sub.-2] ... [S.sub.-j] ... [S.sub.+n]
normalized minimizing
(alternatives 'minuses')
indices of the alternative
Significance of the [Q.sub.2] ... [Q.sub.j] ... [Q.sub.n]
alternative
Priority of the [P.sub.2] ... [P.sub.j] ... [P.sub.n]
alternative
Utility degree of the [N.sub.2] ... [N.sub.j] ... [N.sub.n]
alternative (%)
* The sign [z.sub.i] (+ (-)) indicates that a greater (less) criterion
value corresponds to a greater significance for stakeholders
